The neurobiology of thalamic amnesia: Contributions of medial thalamus and prefrontal cortex to delayed conditional discrimination

G Model

ARTICLE IN PRESS

NBR-2117; No. of Pages 14

Neuroscience and Biobehavioral Reviews xxx (2015) xxx–xxx

Contents lists available at ScienceDirect

Neuroscience and Biobehavioral Reviews journal homepage: www.elsevier.com/locate/neubiorev

Review

The neurobiology of thalamic amnesia: Contributions of medial thalamus and prefrontal cortex to delayed conditional discrimination Robert G. Mair ∗ , Rikki L.A. Miller, Benjamin A. Wormwood, Miranda J. Francoeur, Kristen D. Onos, Brett M. Gibson. Department of Psychology, University of New Hampshire, Durham, NH 03824, United States

a r t i c l e

i n f o

Article history: Received 30 May 2014 Received in revised form 18 December 2014 Accepted 12 January 2015 Available online xxx Keywords: Prefrontal cortex Mediodorsal nucleus Intralaminar nuclei DNMTP Thalamic amnesia Conditional discrimination

a b s t r a c t Although medial thalamus is well established as a site of pathology associated with global amnesia, there is uncertainty about which structures are critical and how they affect memory function. Evidence from human and animal research suggests that damage to the mammillothalamic tract and the anterior, mediodorsal (MD), midline (M), and intralaminar (IL) nuclei contribute to different signs of thalamic amnesia. Here we focus on MD and the adjacent M and IL nuclei, structures identified in animal studies as critical nodes in prefrontal cortex (PFC)-related pathways that are necessary for delayed conditional discrimination. Recordings of PFC neurons in rats performing a dynamic delayed non-matching-to position (DNMTP) task revealed discrete populations encoding information related to planning, execution, and outcome of DNMTP-related actions and delay-related activity signaling previous reinforcement. Parallel studies recording the activity of MD and IL neurons and examining the effects of unilateral thalamic inactivation on the responses of PFC neurons demonstrated a close coupling of central thalamic and PFC neurons responding to diverse aspects of DNMTP and provide evidence that thalamus interacts with PFC neurons to give rise to complex goal-directed behavior exemplified by the DNMTP task. © 2015 Elsevier Ltd. All rights reserved.

Contents 1. 2. 3. 4.

5.

6.

Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . The post thiamine deficiency (PTD) model of the Wernicke–Korsakoff syndrome (WKS) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Experimental studies of the PTD model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Medial thalamic systems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.1. The anterior nuclei . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.2. The mediodorsal (MD), midline (M), and rostral intralaminar (IL) nuclei . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.3. Modulatory effects of medial thalamic stimulation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Electrophysiological studies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5.1. Medial prefrontal recording . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5.2. Medial thalamic recording . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5.3. Effects of medial thalamic inactivation on prefrontal activity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Medial thalamus, global amnesia, and executive function . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

1. Introduction Focal lesions in medial thalamus are associated with global amnesia, affecting the ability to remember declarative information

∗ Corresponding author. Tel.: +1 603 862 3198; fax: +1 603 862 4986. E-mail address: [email protected] (R.G. Mair).

00 00 00 00 00 00 00 00 00 00 00 00 00

in multiple sensory modalities both before and after disease onset while sparing other cognitive functions measured by standard intelligence tests. Thalamic amnesia has been observed with tumors (McEntee et al., 1976; Williams and Pennybacker, 1954), strokes (van der Werf et al., 2003; von Cramon et al., 1985), trauma (Squire et al., 1989), and Wernicke–Korsakoff syndrome (WKS), a disease most commonly associated with thiamine deficiency in chronic alcoholics (Malamud and Skillicorn, 1956; Victor et al., 1989).

http://dx.doi.org/10.1016/j.neubiorev.2015.01.011 0149-7634/© 2015 Elsevier Ltd. All rights reserved.

Please cite this article in press as: Mair, R.G., et al., The neurobiology of thalamic amnesia: Contributions of medial thalamus and prefrontal cortex to delayed conditional discrimination. Neurosci. Biobehav. Rev. (2015), http://dx.doi.org/10.1016/j.neubiorev.2015.01.011

G Model NBR-2117; No. of Pages 14 2

ARTICLE IN PRESS R.G. Mair et al. / Neuroscience and Biobehavioral Reviews xxx (2015) xxx–xxx

The neurological basis of thalamic amnesia is uncertain. In a large cohort study of WKS Victor et al. (1989) observed a consistent correlation between amnesia and lesions damaging the medial mediodorsal nucleus (MD) and adjacent medial thalamic nuclei. Although all these cases also had mammillary body lesions, Victor et al. argued that MD was the critical site of pathology based on five other cases who recovered from the acute phase of Wernicke’s disease with substantial mammillary body pathology but without memory impairment and without apparent medial thalamic damage. Contrary evidence has come from several well-studied amnesic cases of WKS with medial thalamic pathology largely sparing MD and with substantial damage to the mammillary bodies, mammillothalamic tracts, and anterior nuclei (Gold and Squire, 2006; Mair et al., 1979; Mayes et al., 1988). Studies of thalamic amnesia resulting from stroke or trauma has also provided conflicting evidence with some cases where the primary pathology appears to involve MD and adjacent medial thalamic nuclei (Gold and Squire, 2006; von Cramon et al., 1985; Squire et al., 1989; Tanji et al., 2003) and others where the primary pathology is ascribed to the mammillothalamic tracts or anterior nuclei or damage to both medial and anterior thalamic systems (Aggleton and Brown, 1999; Carlesimo et al., 2011; Krill and Harper, 2012; Pergola and Suchan, 2013; van der Werf et al., 2003). Here we focus on the potential contributions of MD and adjacent M and IL thalamic nuclei. Clinical studies have associated infarcts damaging these nuclei with impairment of “executive” aspects of memory (Carlesimo et al., 2011; van der Werf et al., 2003). Lesion studies in animal models have confirmed that MD plays a role in memory and decision making (Mitchell and Chakroborty, 2013) and have provided evidence that combined damage to MD, as well as M and IL, can produce global impairment of delayed conditional discrimination consistent with human WKS (Sections 2–4 below). MD, M and IL are higher order thalamic nuclei that receive Class 1/driver and Class 2/modulator inputs from PFC, forming cortico-thalamocortical pathways thought to regulate transmission of information in neural networks involving PFC, parietal, and medial temporal cortices (Saalmann, 2014; Sherman and Guillery, 2002, 2011; Steriade et al., 1997). Thus lesions damaging these nuclei should affect the timing of activity in distributed neural networks thought to play a critical role in memory (Johnson and Knight, 2015; Staudigl et al., 2012; Watrous et al., 2015). Recent electrophysiological studies in our lab have confirmed an important role of central thalamus and PFC in delayed conditional discrimination and provided evidence that the “executive functions” of central thalamus and PFC reflect a fundamental concern with planning, actions, and actionoutcomes that underlie complex intentional behavior (Section 5). 2. The post thiamine deficiency (PTD) model of the Wernicke–Korsakoff syndrome (WKS) The PTD model first demonstrated a consistent correlation between lesions centered on the internal medullary lamina, involving MD and adjacent M and IL nuclei, and a pattern of behavioral impairment consistent with WKS: delayed conditional discrimination impairment in multiple sensory modalities coupled with a spared capacity for rule-based responding in serial reversal learning (Mair, 1994). Clinical studies have attributed WKS to thiamine (vitamin B1) deficiency most frequently in chronic alcoholic persons and associated this syndrome with bilateral lesions along the walls and floor of the third ventricle, periaqueductal areas of the midbrain, and the floor of the fourth ventricle (Malamud and Skillicorn, 1956; Victor et al., 1989), as well as signs of widespread cortical impairment in imaging studies (Pitel et al., 2012; Reed et al., 2003; Sullivan and Pfefferbaum, 2009). Mair et al. (1985, 1988) developed the PTD model to examine chronic effects that outlast a subacute bout of thiamine deficiency,

using existing methods to deplete thiamine rapidly by dietary restriction and daily pyrithiamine injections for about 2 weeks until rats reached a critical stage of impairment when thiamine deficiency was reversed by injection of a large dose of thiamine. This treatment produces a characteristic thalamic lesion centered on the internal medullary lamina, tissue loss in the mammillary bodies, signs of diffuse degeneration of thalamocortical projections, and localized loss of monoamine and amino acid neurotransmitters: pathological effects consistent with human WKS (Fig. 1; Langlais et al., 1987, 1988; Mair et al., 1985, 1988, 1989, 1991a,b). Subsequent studies provided evidence that PTD-induced lesions are caused by an excitotoxic response to elevated levels of extracellular glutamate associated with down regulation of astrocytic glutamate transporters that can be prevented by the noncompetitive NMDA antagonist dizocilpine (Hazell et al., 2001; Langlais and Mair, 1990; Langlais and Zhang, 1993; Robinson and Mair, 1992; Zhang et al., 1995). This apparent excitotoxic reaction occurs in the context of multiple metabolic and vascular consequences of thiamine deficiency related to oxidative stress, excitotoxicity, and cerebral inflammation (Hazell and Butterworth, 2009). Behavioral analyses using directly comparable tasks have revealed a pattern of impairment in the PTD model consistent with human WKS (Fig. 2) that persists chronically for many months after restitution of nutritional status (Mair et al., 1988). Comparative neuropsychological studies have shown that WKS affects delayed conditional discriminations of sensory stimuli in multiple sensory modalities, including delayed matching (DM) or nonmatching (DNM) tasks in which subjects choose between two response alternatives based on stimulus information presented in a preceding sample trial (Aggleton et al., 1988; Oscar-Berman et al., 1992; Squire et al., 1988). WKS patients also make more errors to criteria learning two choice discrimination and serial reversal learning tasks, although they retain the ability to perform discriminations at criteria once learned and exhibit positive transfer across a series of reversal problems (Oscar-Berman and Zola-Morgan, 1980). PTD rats are likewise impaired in multiple sensory modalities performing two choice DM and DNM tasks (Knoth and Mair, 1991; Mair et al., 1985, 1988, 1991a; Mumby et al., 1995, 1999; Robinson and Mair, 1992) and in errors to criterion in discrimination and serial reversal learning (Mair et al., 1991b). Like WKS patients, PTD rats exhibit a spared capacity for rule-based responding shown by the abilities to perform discriminations at criterion once they have been learned and positive transfer across a series of reversal discrimination problems (Fig. 2). It should be noted that while these tasks allow for direct comparison of animal and human performance for functions spared and impaired by amnesia, they by no means represent the full range of cognitive functions affected by WKS (Butters and Cermak, 1980; Fama et al., 2012; Oscar-Berman, 2012; Talland, 1965). 3. Experimental studies of the PTD model The PTD rat has three main claims to validity as a model of WKS: etiology associated with thiamine deficiency, symmetrical lesions involving mammillary bodies and medial thalamus and signs of broad thalamo-cortical degeneration (Fig. 1), and similar patterns of chronic behavioral impairment (Fig. 2). Both PTD and WKS have also been associated with lesions involving multiple systems in the brain including signs of diffuse cortical pathology that limit their utility for elucidating the contributions of specific nuclei or pathways to signs of memory impairment. To determine the critical sites of pathology in the PTD model, we compared the effects of lesions selectively targeting different systems affected by PTD treatment. We found that medial thalamic lesions involving both MD and rostral IL, produced by either electrolytic radio frequency current or microinjection of excitotoxic

Please cite this article in press as: Mair, R.G., et al., The neurobiology of thalamic amnesia: Contributions of medial thalamus and prefrontal cortex to delayed conditional discrimination. Neurosci. Biobehav. Rev. (2015), http://dx.doi.org/10.1016/j.neubiorev.2015.01.011

G Model NBR-2117; No. of Pages 14

ARTICLE IN PRESS R.G. Mair et al. / Neuroscience and Biobehavioral Reviews xxx (2015) xxx–xxx

3

Fig. 1. Low power micrographs showing characteristic damage produced by PTD treatment. A–C are stained with luxol fast blue, hematoxylin, and eosin and D with the Fink–Heimer method to demonstrate degenerating axon terminals in frontal cortex (materials from Mair et al., 1988, 1989, 1991a). Coronal (A) and sagittal (B) sections reveal bilaterally symmetrical lesions thalamic lesions (arrows) posterior to anterior from fasciculus retroflexus to the mammillothalamic tract, centered on the internal medullary lamina and involving rostral intralaminar and adjacent areas of the mediodorsal nucleus. A more posterior coronal section (C) shows loss of tissue within the mammillary bodies and the resulting expansion of the mammillary recess of the third ventricle (arrow). Fink–Heimer stained material (D) shows diffuse terminal degeneration in superficial layers and heavier degeneration in middle layers throughout frontal cortex except for ventral medial prefrontal cortex (arrows). MD, mediodorsal nucleus; AM, anteromedial nucleus; mt, mammillothalamic tract; fr, fasciculus retroflexus.

drugs, produce patterns of behavioral impairment comparable to PTD treatment (Fig. 2; reviewed below). By contrast, lesions damaging adjacent thalamic nuclei, including the anterior (Mair et al., 2003), midline (Hembrook and Mair, 2011; Mair and Lacourse, 1992), or ventromedial (Bailey and Mair, 2005; Burk and Mair, 1999) nuclei produce more limited deficits that are not as broad in scope as the PTD model. Other studies have shown that lesions damaging fornix or mammillary bodies (Mair and Lacourse, 1992; Young et al., 1996) or depleting neurotransmitters beyond deficits produced by PTD treatment (Koger and Mair, 1992) are also not associated with behavioral impairments comparable to the effects of PTD treatment or MD and IL lesions. The implication of MD and IL as a critical site of pathology in the PTD model is consistent with evidence that bilateral damage to these nuclei predicts delayed conditional discrimination impairment following PTD treatment and that dizocilpine protects from these lesions and from signs of behavioral impairment. Recent work by Savage and colleagues have provided evidence that PTD treatment has broad effects on the activity of hippocampus and cerebral cortex, associated with loss of cholinergic innervation, that may contribute to the direct effects of MD and IL pathology on behavioral deficits in the PTD model (Savage et al., 2012). 4. Medial thalamic systems 4.1. The anterior nuclei Uncertainty remains about the precise contributions of systems within medial thalamus affected by PTD-induced lesions. Medial thalamus is a tangle of juxtaposed nuclei and fiber tracts that make it difficult to damage one substantially without affecting others. Clinical studies have emphasized the importance of damage to the

mammillary bodies, mammillothalamic tract (MTT), and anterior thalamic nuclei (AT) in diencephalic amnesia (Aggleton and Brown, 2006; van der Werf et al., 2003). An obvious attraction of this idea is the direct link of this circuit to hippocampus and the implication of hippocampal dysfunction as a unitary cause of both thalamic and medial temporal lobe amnesia. It is well established that AT lesions in rats affect T-maze, plus-maze, and radial arm maze measures of allocentric spatial memory that depend on hippocampal function (Aggleton et al., 1996; Aggleton and Nelson, 2014; Dumont and Aggleton, 2013; Mair et al., 2003; Savage et al., 2011; Sziklas and Petrides, 2007). Indeed, Mitchell and Dalyrmple-Alford (2005, 2006) have provided direct evidence that AT lesions have greater effects than IL lesions on radial maze spatial memory tasks (but see Mair et al., 1998). Other results suggest that AT lesions have a circumscribed effect on memory that does not extend globally to all sensory modalities. AT lesions are reported to spare recognition or recency judgments of objects or odors, except for recency judgments made within the same session in which stimuli are initially presented (Dumont and Aggleton, 2013; Wolff et al., 2006). AT lesions have also been found to spare response-related (egocentric spatial) memory (Mitchell and Dalyrmple-Alford, 2005, 2006). AT lesions affect a range of spatial behaviors consistent with the involvement of these nuclei in pathways that represent allocentric spatial information (Aggleton and Nelson, 2014). A review of this literature is beyond the scope of this paper. The pronounced effects of AT lesions on allocentric spatial memory seems consistent with the emerging view that individual nuclei in medial thalamus contribute to distinct aspects of memory. Zhang et al. (1995) observed neurocytological pathology in the anterior ventral and interanteromedial nuclei during early stages of PTD treatment. The early onset of these lesions, coupled with the

Please cite this article in press as: Mair, R.G., et al., The neurobiology of thalamic amnesia: Contributions of medial thalamus and prefrontal cortex to delayed conditional discrimination. Neurosci. Biobehav. Rev. (2015), http://dx.doi.org/10.1016/j.neubiorev.2015.01.011

G Model NBR-2117; No. of Pages 14 4

ARTICLE IN PRESS R.G. Mair et al. / Neuroscience and Biobehavioral Reviews xxx (2015) xxx–xxx

Fig. 2. Comparison of behavioral impairment in human Wernicke–Korsakoff patients (A, B; replotted from Oscar-Berman and Bonner, 1989 and Oscar-Berman and ZolaMorgan, 1980), PTD treated rats (C, D; replotted from Robinson and Mair, 1992 and Mair et al., 1991a,b), and rats with radiofrequency thalamic lesions of the lateral internal medullary lamina (E, F; replotted from Mair and Lacourse, 1992 and Harrison and Mair, 1996). These groups showed delay independent impairments of delayed non-matching to sample tasks (A, C, E), and for serial reversal learning (B, D, F) increased errors to criterion with a spared capacity for rule-based responding demonstrated by the ability to perform discriminations at criterion once learned and to exhibit positive transfer across serial reversal problems (B, D, F).

well-established effects of AT lesions on hippocampal-dependent memory, must be taken as evidence that AT pathology contributes to the behavioral impairment of the PTD model. The available evidence, however, suggests that AT pathology is at best only part of the story. Experimental thiamine deficiency produces extensive damage to MD, IL, and M nuclei (Fig. 1) that occurs later in the progression of this treatment (Zhang et al., 1995). The behavioral impairments that link the PTD model to WKS (Fig. 2) are only observed when thiamine deficiency is reversed later in rats that exhibit extensive internal medullary lamina lesions involving MD, M, and IL (Langlais et al., 1992; Mair et al., 1988, 1991a,b; Mumby et al., 1995, 1999; Robinson and Mair, 1992). Other reports have shown that lesions restricted to anterior areas of thalamus or directly damaging fornix or mammillary bodies do not produce behavioral impairments comparable to PTD (Mair and Lacourse, 1992; Mair et al., 1992). 4.2. The mediodorsal (MD), midline (M), and rostral intralaminar (IL) nuclei Thalamic amnesia has also been ascribed to MD pathology, alone or in combination with damage to adjacent M and IL nuclei or MTT

circuits involving the AT (Mitchell and Chakroborty, 2013; Pergola and Suchan, 2013; van der Werf et al., 2003; Victor et al., 1989). Lesion studies in the rat have also shown that damage to MD, IL, or M nuclei can impair memory function independent of or dissociable from the effects of MTT or AT damage (Bailey and Mair, 2005; Hembrook and Mair, 2011; Mitchell and Dalyrmple-Alford, 2005, 2006). Discrete experimental lesions of MD have been found to produce delay-dependent impairment of decision making in delayed matching to position (DMTP) and DNMTP (Bailey and Mair, 2005; Young et al., 1996) and other delayed conditional discriminations (Kessler et al., 1982; Zola-Morgan and Squire, 1985), sparing performance at short memory delays. By contrast PFC lesions (Mair et al., 1998; Porter et al., 2000; Young et al., 1996) produce delay-independent impairment of these tasks, as do lesions of ventral striato-pallidal pathways that connect PFC with medial thalamus (Fig. 3; Burk and Mair, 2001; Mair et al., 2002; Zhang et al., 2005). This suggests that the influence of MD on PFC function increases as the length of the memory delay increases in these short-term memory tasks. There are several mechanisms that might explain this delay dependence. First, MD lesions could interfere with reentrant activation between MD and PFC thought to maintain sustained delay period activity

Please cite this article in press as: Mair, R.G., et al., The neurobiology of thalamic amnesia: Contributions of medial thalamus and prefrontal cortex to delayed conditional discrimination. Neurosci. Biobehav. Rev. (2015), http://dx.doi.org/10.1016/j.neubiorev.2015.01.011

G Model NBR-2117; No. of Pages 14

ARTICLE IN PRESS R.G. Mair et al. / Neuroscience and Biobehavioral Reviews xxx (2015) xxx–xxx

5

Fig. 3. Delayed matching-to-position impairment in rats following lesions in medial thalamus (replotted from Bailey and Mair, 2005), medial prefrontal lesions (replotted from Mair et al., 1998), ventral striatum (replotted from Burk and Mair, 2001), and ventral pallidum (replotted from Zhang et al., 2005). These results show delay-independent impairment for lesions damaging medial prefrontal cortex and related ventral striato-pallido-thalamic pathways. Only the MD was associated with impairment that increased significantly at longer delays.

and hypothesized to function as a short-term active memory store (Alexander and Fuster, 1973; Watanabe and Funahashi, 2012). Second, MD receives input from limbic-related memory systems that may provide a source of remembered information or limbic-related activation (Bailey and Mair, 2005; Mitchell and Chakroborty, 2013). Indeed, evidence has been presented that PFC depends on hippocampus for rapid associative learning (Euston et al., 2012). The importance of MD as a source of limbic-related input to PFC is supported by recent work with the object-in-place task. This measure of associative recognition memory has been shown to depend on interactions between MD and PFC (Cross et al., 2012), as well as hippocampus-PFC and hippocampus-perirhinal cortex circuits (Barker and Warburton, 2013). Finally, MD pathology could affect top-down control of memory consolidation and retrieval by PFC (Euston et al., 2012). Lesion studies have also provided evidence that MD plays a fundamental role in instrumental decision-making, affecting reward-based and cue-guided response selection. In monkeys, medial MD lesions have been shown to impair the ability to adjust decision-making when pairs of two stimuli are associated with distinct reinforcements and the reinforcer value of one is reduced by selective satiation (Mitchell et al., 2007). Similar deficits have been observed with lesions of amygdala (Malkova et al., 1997), orbital PFC (Izquierdo et al., 2004), or disconnection of MD with amygdala and orbital PFC (Izquierdo and Murray, 2010). These results suggest that in monkeys pathways involving MD and amygdala and orbital prefrontal cortex contribute to reward-based decision making (Izquierdo and Murray, 2010). Corbit et al. (2003) likewise report that MD lesions eliminate the effects of outcome devaluation in the rat. Here rats were trained to press two levers, each associated with a distinct reinforcement. One of these reinforcements was then devalued by selective extinction and preference for pressing the two levers tested in extinction. Normally rats exhibit a preference for pressing the lever associated with the reinforce that has not been devalued. Rats with

MD lesions exhibited a nonselective decrease in responding to both levers. Later work showed that this effect is eliminated when MD is lesioned after instrumental training associating each lever with a distinctive reward (Ostlund and Balleine, 2005), suggesting that MD is important for the acquisition, but not the expression of action-outcome contingencies. This distinction is consistent with evidence that prelimbic PFC lesions affect reinforcement devaluation when rats are lesioned before training (Corbit and Balleine, 2003). More recently Bradfield et al. (2013) demonstrated similar impairment of reinforcement devaluation with crossed unilateral lesions of MD and prelimbic cortex (combined with a corpus callosum lesion to prevent inter-hemispheric transfer). None of these lesions affected outcome-induced reinstatement in which presentation of one of the reinforcers in extinction selectively increases responding on its associated lever (Corbit and Balleine, 2003; Corbit et al., 2003; Bradfield et al., 2013). Taken together, these results suggest that pathways connecting MD and prelimbic PFC contribute to the ability to learn action-outcome contingencies to guide goal directed responding. Ostlund and Balleine (2007, 2008) have shown that lesions damaging MD or its projection areas in orbital PFC interfere with cue-guided action selection in Pavlovian-instrumental transfer. Here cues associated with a particular reward normally facilitate responding on an action previously paired with that reward when tested in extinction. Rats with MD lesions failed to show cue-specific response selection, although they exhibited normal outcome-guided response selection, increasing responses of an action when its associated reward was presented noncontingently during extinction. Others have reported that both MD (Chudasama et al., 2001) and PFC (Bussey et al., 1997) lesions spare cue-guided choice during initial acquisition of a visual discrimination, while increasing errors to criterion when contingencies are reversed: impairments interpreted as evidence of impaired stimulus-response learning. These studies of decision-making in rats with selective MD and PFC lesions suggest that circuits

Please cite this article in press as: Mair, R.G., et al., The neurobiology of thalamic amnesia: Contributions of medial thalamus and prefrontal cortex to delayed conditional discrimination. Neurosci. Biobehav. Rev. (2015), http://dx.doi.org/10.1016/j.neubiorev.2015.01.011

G Model NBR-2117; No. of Pages 14 6

ARTICLE IN PRESS R.G. Mair et al. / Neuroscience and Biobehavioral Reviews xxx (2015) xxx–xxx

involving these areas support the acquisition of goal-directed actions, affecting encoding of action-outcome, stimulus-outcome, and stimulus-response relationships. This raises the possibility that the deficits in basic aspects of goal-directed responding may underlie the effects of MD and PFC lesions on DM and DNM tasks, where correct selection between response alternatives requires a conditional decision based on trial-specific information from a preceding sample response. The ability of rats with MD and PFC lesions to perform simple discriminations once learned demonstrates that these areas are not needed to decide between response alternatives based on a well-established stimulus-response rule. Convergent evidence suggests that the effects of MD lesions on memory are exacerbated when pathology extends to include substantial portions of the adjacent M and IL nuclei. IL damage has been associated with delay-independent effects on DMTP and DNMTP, affecting performance at short delays spared by MD lesions but affected by PFC lesions (Bailey and Mair, 2005; Burk and Mair, 1998; Young et al., 1996) and consistent with the PTD model and human WKS (Fig. 2). Other studies have provided evidence that IL lesions affect other memory tasks that are spared by MD or even PFC lesions. For instance, IL lesions produce persistent impairment of go/no go responding in continuous olfactory DNM, while sparing go-no go responding for simple olfactory discrimination—a pattern observed with pyriform cortex lesions, but not MD, PFC, hippocampus, or entorhinal cortical lesions (Koger and Mair, 1994; Mair et al., 1998; Zhang et al., 1998). Similarly, IL lesions have been shown to affect DNMTP where choices were drawn at random from eight arms of a radial maze to prevent a response- (egocentric-) mediated solution—a task affected by lesions of hippocampus and AT but not PFC (Mair et al., 1998, 2003; Porter et al., 2000). M and IL nuclei participate in cortico-thalamo-cortical pathways projecting widely to superficial layers of prefrontal, frontal, parietal, and medial temporal cortices regions as well as to anatomically related areas of the basal ganglia (Groenewegen and Berendse, 1994; Hoover and Vertes, 2012; Smith et al., 2014; Vertes et al., 2012). The broad effects of IL lesions seem consistent with these widely distributed modulatory projections. The combination of IL and MD pathology, almost inevitably associated with PTD and substantial IL lesions, would affect both the core projections of MD to PFC and the more diffuse superficial matrix projections of the MD and IL to prefrontal, frontal, and parietal areas of cortex (Jones, 2009). This is a combination that should have broad effects on thalamic control of cortical synchrony required for normal memory and other cognitive functions (León-Domínguez et al., 2013; Saalmann, 2014). The ability of rats with substantial MD, IL, and prefrontal lesions to perform simple discriminations (once learned) with similar choice responses and reinforcement contingencies (Burk and Mair, 1998; Harrison and Mair, 1996; Koger and Mair, 1994; Mair et al., 1991a,b; Zhang et al., 1998) demonstrate a preserved capacity for simple decision making where the correct choice is associated consistently with a particular location or sensory stimulus. 4.3. Modulatory effects of medial thalamic stimulation In addition to cortical and basal ganglia-related afferents, the M and IL nuclei receive visceral and arousal-related inputs from brainstem areas and from histaminergic and orexinergic neurons in hypothalamus (Steriade et al., 1997; Bayer et al., 2002; Krout et al., 2002). These nuclei thus seem organized to modulate the activity of broadly distributed cortical and basal ganglia circuits in conjunction with behavior state (Haber and Calzavara, 2009; Jones, 2009; Schiff, 2008; van der Werf et al., 2002). Recordings of neuronal activity in primates have revealed precisely timed activity of neurons in IL and paralaminar regions of central thalamus related to behaviorally significant events (Matsumoto et al., 2001; Minamimoto and Kimura, 2002; Schiff et al., 2013; Wyder et al.,

2003). Functional imaging has similarly shown activation of central thalamus during alert, wakeful states (Paus, 2000), particularly when demands are increased on attention or working memory (Burianova and Grady, 2007; Chee and Choo, 2004; Kinomura et al., 1997). Clinical studies have identified central thalamus as a location of lesions associated with disorders of arousal and impairments of consciousness in human patients (Schiff, 2008). Lesion studies in rats have demonstrated significant effects of central thalamic lesions on open field exploration (Mair et al., 1991a,b; Mair and Lacourse, 1992) and response speed (Burk and Mair, 1998; Mair and Lacourse, 1992; Robinson and Mair, 1992) indicative of diminished arousal. Indeed, diminished arousal and blunted affect are frequently observed WKS (Talland, 1965; Mair, 1994; Oscar-Berman, 2012) and thalamic infarct amnesia (Schmanmann, 2003; van der Werf et al., 2003). Central thalamus has been proposed as a potential site for activating cognitive function based on the evidence that it serves as a hub coordinating the activity of distributed cortical-basal ganglia circuits that give rise to organized actions (Shah and Schiff, 2012; Schiff, 2008). To test this possibility, Mair and Hembrook (2008) conducted a series of experiments manipulating the activity of central thalamus with microinjected drugs and electrical deep brain stimulation (DBS) and examining the effects on DMTP. They found that inhibiting activity with the GABAA agonist muscimol produces dose-dependent, delay-independent impairment consistent with the effects of IL lesions (Fig. 4). The localization of this effect was confirmed by comparing the effects of muscimol infusions in an anatomical control site 1.5 mm above the experimental site and by comparing the effects of reversible inactivation with lesions in the area of the injection site in previous studies (Burk and Mair, 1998; Bailey and Mair, 2005). By contrast, reducing GABAA tone with the ␤-carboline FG-7142 enhanced performance at low doses and impaired performance at higher doses (Fig. 4). Similarly orexin-A enhanced activity at the lowest dose tested, while the cholinergic agonist carbachol impaired performance at the highest dose tested (Mair and Hembrook, 2008). Taken together these results suggest that inhibiting central thalamic activity impairs DMTP performance (like central thalamic lesions), while treatments increasing activity can enhance performance at low doses and impair performances at higher doses, following an inverted-U (Yerkes-Dodson) dose-response curve. The effects of DBS were consistent with these results. Performance was enhanced at current levels near threshold for activating neurons (generally ≤0.01 mA) and impaired at higher currents (≥0.10 mA, below levels having observable effects on responding). DBS results also revealed evidence of temporal specificity. DBS was delivered for 1.0 s trains (of 0.2 ms pulses@120 Hz) at different times during DMTP trials. Stimulation affected performance significantly when delivered at the start of the memory delay and choice response periods, but not at the start of trials or sample response periods (Fig. 4). More recently, DBS targeting nucleus reuniens in ventral midline thalamus was found to impair DNMTP performance when applied at 0.10 mA for a similar 1.0 s train at the start of delay and choice, but not initiation or sample periods. Interestingly this same stimulation had no effect when delivered with the same timing for an identical choice response in a simple discrimination task (Hembrook et al., 2011). Thus the disruptive effects of this stimulation seem specific to complex aspects of conditional discrimination, a result consistent with the differential effects of medial thalamic lesions on conditional discrimination and simple discrimination tasks (reviewed above). 5. Electrophysiological studies Lesions studies suggest that medial thalamic-PFC pathways are important for complex decision-making required for delayed

Please cite this article in press as: Mair, R.G., et al., The neurobiology of thalamic amnesia: Contributions of medial thalamus and prefrontal cortex to delayed conditional discrimination. Neurosci. Biobehav. Rev. (2015), http://dx.doi.org/10.1016/j.neubiorev.2015.01.011

G Model NBR-2117; No. of Pages 14

ARTICLE IN PRESS R.G. Mair et al. / Neuroscience and Biobehavioral Reviews xxx (2015) xxx–xxx

7

conditional discrimination, but not for simple rule-based responding in sensory discriminations once they have been learned. This distinction provides evidence that the effects of MD and IL lesions on conditional discrimination do not result from global debilitation or impairments in sensory, motor, reinforcement, or motivational functions required for well-trained discrimination tasks. Studies of instrumental decision-making provide evidence that MD and PFC support the ability to adjust decision making in response to reward devaluation, to cues associated with anticipated outcomes in pavlovian instrumental transfer, and to changed stimulus-reward contingencies in reversal learning. We are conducting several lines of electrophysiological investigation to elucidate the contributions of medial thalamus and PFC to delayed conditional discrimination. By studying the activity of neurons in rats performing a flexible DNMTP task we want to understand the underlying processes supported by medial thalamus and PFC that contribute to instrumental decision making in general and to delayed conditional discrimination in particular. To understand the influence of spatial and temporal context on the response properties of thalamic and prefrontal neurons we are carrying out simultaneous analyses of activity correlated with multiple events during sample and choice phases of DNMTP and with spatial location and movement in videotracking records. In each of these studies, neuronal activity is recorded with moveable arrays of tetrodes as rats perform a dynamic DNMTP task trained in an open octagonal arena with numerous visible cues (Fig. 5). Recorded activity is processed offline to identify signals from isolated neurons, which are then correlated with behavioral events as peri-event time histograms (PETH) and raster plots. To analyze effects of spatial context, all sessions are videotaped and head direction and location mapped using LEDs attached to the recording headstage. These data are then correlated with electrophysiological recordings to map the relationships between neural activity spatial location (place fields), head direction, and movement vectors. 5.1. Medial prefrontal recording

Fig. 4. Effects of manipulating medial thalamic activity on delayed matching to position. (DMTP; replotted from Mair and Hembrook, 2008). Inhibition with the GABAA agonist muscimol produced dose dependent impairment that was not observed with injections at an anatomical control site (Anat Cnt). Activation with the inverse benzodiazepine agonist FG-7142 enhanced performance at low doses and impaired performance at higher doses, following an inverted-U dose-response function. Excitation with 1 s trains of electrical deep brain stimulation also produced signs of an inverted-U relationship between the level of stimulating current and DMTP performance. Here there were signs of temporal specificity. DBS coinciding with the delay and choice lever presses had significant effects on performance while DBS delivered earlier in trials, at the time of the initiate (start) or sample lever presses did not.

We recorded activity from 900 neurons in medial PFC of 5 rats from frontal area 2, through anterior cingulate and prelimbic areas to ventral infralimbic cortex (Onos et al., 2013; Francoeur et al., 2014). We identified 10 distinct response types, characterized by precisely timed event-related activity, that account for 95% of the 288 neurons that had statistically significant PETH responses (see Fig. 6 for examples). These response types are consistent with previous reports that have linked prefrontal activity with planning (or preparation) ending prior to the start response (Gallivan et al., 2013; Totah et al., 2009), execution of specific actions (Euston et al., 2012; Jung et al., 1998; Poucet, 1997), reinforcement (Alexander and Brown, 2011; Pratt and Mizumori, 2001; Strait et al., 2014), and delay periods between the sample and choice responses (Chang et al., 2002; Cowen and McNaughton, 2007; Horst and Laubach, 2012). Our study has identified all of these response types in a single task and revealed precise timing of different response types characterizing the temporal context of actions and events that constitute DNMTP trials. Action related responses include neurons responding consistently during lever pressing and movement, as well as more specifically to lever presses that demarcate the start of sample and choice phases and movements that encode specific sample and choice responses. Reinforcement-related responses include neurons that fire in anticipation of impending reinforcement, neurons with activity time-locked to the delivery of reinforcement, and others that respond following errors when reinforcement is not delivered. Anticipatory responses start when rats begin to move from the base lever, following start or delay responses,

Please cite this article in press as: Mair, R.G., et al., The neurobiology of thalamic amnesia: Contributions of medial thalamus and prefrontal cortex to delayed conditional discrimination. Neurosci. Biobehav. Rev. (2015), http://dx.doi.org/10.1016/j.neubiorev.2015.01.011

G Model NBR-2117; No. of Pages 14 8

ARTICLE IN PRESS R.G. Mair et al. / Neuroscience and Biobehavioral Reviews xxx (2015) xxx–xxx

Fig. 5. Schematic representation of the DNMTP task. Rats were trained in an octagonal arena, 62 cm in diameter, with four retractable levers centered on walls 90◦ apart. Each lever had a drinking spout and a stimulus light above to deliver and signal delivery of water reinforcement, two drops of water delivered 1 s apart by activation of a miniature solenoid valve. The arena was located in a Faraday cage with a screen door that provided ambient illumination and many visible external cues. Each DNMTP trial includes a sequence of four lever presses (start, sample, delay, and choice) involving three of the four levers in the chamber. The base lever extended to begin each trial is randomly selected, making the start location unpredictable and preventing rats from relying on a fixed stimulus-response strategy to identify the correct choice. The base lever is retracted following the start press, causing a lever 90◦ to the left or right (randomly selected) to extend for the sample. The sample lever press delivers reinforcement to the spout immediately above the sample lever (signaled by the panel light) and causes the sample lever to retract and the base lever to extend again for the delay response. This retracts with the first press after the memory delay ends causing the levers 90◦ to the left and right to extend for the choice. The trial ends and reinforcement is delivered when the choice lever that was not extended for the sample (i.e. non-match to sample) is pressed. DNMTP is trained in daily sessions for a maximum of 60 trials or 60 m. Asterisks indicate lever presses associated with reinforcement.

and end after reinforcement events that follow sample or correct choice responses or immediately after incorrect choice lever presses when reinforcement is not delivered. These seem consistent with the effects of MD and prefrontal lesions on the acquisition action-outcome contingencies where lesioned animals are impaired adjusting decision making when anticipated reinforcers are devalued (Corbit and Balleine, 2003; Corbit et al., 2003; Bradfield et al., 2013). As in previous reports (Chang et al., 2002; Cowen and McNaughton, 2007; Horst and Laubach, 2012), delayrelated activity was not as prominent in rat PFC as in earlier descriptions of primates (Fuster and Alexander, 1971; Funahashi, 2006). Anatomical analyses showed evidence of functional specialization with localized distributions of responses related to planning, delay, and reinforcement anticipation. Each of these localizations is consistent with information provided by cortico-cortical and subcortical inputs (Heidbreder and Groenewegen, 2003; Hoover and Vertes, 2007) and with behavioral effects of lesions. Planning (preparatory) activity were localized in FR2, AC, and dorsal PL cortices, areas with more robust connections to non-limbic sensorimotor areas and lesions that specifically increase the time taken to initiate learned action sequences (Bailey and Mair, 2004, 2007; Kesner and Churchwell, 2011). Delay-related responses were concentrated in ventral prelimbic and infralimbic cortices; areas innervated by CA1/subiculum and a location of lesions with profound effects on delayed response tasks (Chudasama, 2011; Dalley et al., 2004; Kesner and Churchwell, 2011). Reinforcement anticipation responses were localized in anterior prelimbic cortex, an area innervated by basolateral amygdala and ventral pallidum and the location of lesions that affect encoding of action-outcome contingency (Balleine and Dickinson, 1998; Bradfield et al., 2013; Corbit and Balleine, 2003). Other responses related to actions and action outcomes were more broadly dispersed: distributions that seemingly represent a common core of information shared across medial PFC. Taken together these anatomical mappings provide evidence of functional specialization of local PFC areas consistent with cortico-cortical and subcortical inputs as well as shared information related to actions and action-outcomes that suggest commonalities between functions supported by different areas of medial prefrontal cortex. Subareas of medial prefrontal cortex have extensive interconnections that provide a likely mechanism for transmitting common information about actions and outcomes that we have found to be widely distributed across these regions (Heidbreder and Groenewegen, 2003; Hoover and Vertes, 2007). Place analyses revealed spatial patterns of activation consistent with PETH results. For instance, movement-related responses showed maximal activation along paths running directly between

levers and delay-related responses were between levers pressed for the sample response and adjacent levers that were pressed at the end of the delay phase (Fig. 6). Fig. 6 shows examples of rasters, PETHs, and place fields for prefrontal neurons responding during movement between levers and during the memory delay. A subset of neurons had place fields restricted to a circumscribed area of the arena, representing information about specific behavioral events and their spatial context. The preponderance of neuronal responses related to actions and action outcomes seems consistent with results of lesion studies implicating PFC in encoding action-outcome associations that underlie the acquisition of goal-directed behavior (Balleine and Dickinson, 1998; Bradfield et al., 2013; Corbit and Balleine, 2003). The finding of responses related to diverse aspects of DMNTP and their spatial and temporal context confirms the importance of PFC in delayed conditional discrimination and seems consistent with findings of delay-independent impairment in rats with PFC lesions extending to trials with minimal memory delays (Harrison and Mair, 1996; Mair et al., 1998; Stevens and Mair, 1998; Young et al., 1996). Individual trials in these tasks involve a coordinated sequence of actions encompassing sample and choice responses that place demands on planning and temporal organization of actions. Correct choices are not defined directly by a particular stimulus or response, as in discrimination or response learning, but must be determined flexibly based on the preceding sample response—in the context of a left sample response, a right choice is correct and vice versa. This creates a difficult decisional process that inevitably results in errors and almost certainly benefits from the ability to anticipate and monitor the outcomes of actions. The preserved capacity to perform simple discriminations once learned in rats with PFC lesions sufficient to impair conditional discrimination (Harrison and Mair, 1996; Koger and Mair, 1994; Porter et al., 2000; Stevens and Mair, 1998) provides evidence that these functions of prefrontal neurons are not necessary for rule-based responding where a correct response can be determined by an invariable stimulus-response association. 5.2. Medial thalamic recording The MD, M, and IL nuclei receive both driver (class 1) and modulatory (class 2) inputs from PFC. While they all provide diffusely projecting matrix projections to superficial layers of PFC and related cortical areas, MD is the main source of precisely projecting core input to middle cortical layers of PFC. Thus MD, M, and IL are important nodes in transthalamic circuitry implicated in selective modulation and gating of cortical activity and dynamic coupling of cortical areas during complex cognitive tasks such as DNMTP (Bay

Please cite this article in press as: Mair, R.G., et al., The neurobiology of thalamic amnesia: Contributions of medial thalamus and prefrontal cortex to delayed conditional discrimination. Neurosci. Biobehav. Rev. (2015), http://dx.doi.org/10.1016/j.neubiorev.2015.01.011

G Model NBR-2117; No. of Pages 14

ARTICLE IN PRESS R.G. Mair et al. / Neuroscience and Biobehavioral Reviews xxx (2015) xxx–xxx

9

Fig. 6. A comparison of event related responses in medial prefrontal cortex (PFC; panels A, C, E, G) and the mediodorsal thalamic nucleus (MD; panels B, D, F, H) in rats performing a dynamic delayed non-matching-to-position (DNMTP) task (data from Francoeur et al., 2014; Miller et al., 2014; Onos et al., 2013). Peri-event time histograms (PETH) and rasters provide examples of the similarity in timing of PFC and MD responses and the tendency for MD neurons to exhibit higher levels of baseline activity during training sessions and less distinct event-related responses. A and B show movement-related activity preceding each of the lever press responses. Note that movement increases immediately after start and delay responses and is delayed after sample and choice responses when rats received reinforcement from a spout immediately above the response lever (two 0.1 s pulses of water, 1.0 s apart). C and D show typical delay period activity. We have found that delay period activity for this task corresponds to the delivery of reinforcement following sample and correct choice responses (both of which are reinforced), but not choice errors (which do not deliver reinforcement). PETH responses are scaled (from 0) for each neuron to represent average activity/trial so that relatively rare error responses can be compared directly to sample and correct choice responses. Place maps show the spatial distribution of activity (red—high, blue—low) during DNMTP training. Both PFC (E) and MD (F) neurons exhibiting movement-related activity (A, B) show higher levels of activity along pathways between response levers. Similarly, both PFC (G) and MD (H) exhibiting delay-related activity showed higher levels of activity in the vicinity of base levers (marked with white B’s) where start and delay responses were made and lower levels in the vicinity of levers where sample and choice responses were made. The location of base levers was changed randomly between either two (N, S or E, W) or four (N, E, S, W) locations on each trial. Sample and choice responses were made to levers 90◦ to the left or right of the base lever. (For interpretation of the references to color in figure legend, the reader is referred to the web version of the article.)

Please cite this article in press as: Mair, R.G., et al., The neurobiology of thalamic amnesia: Contributions of medial thalamus and prefrontal cortex to delayed conditional discrimination. Neurosci. Biobehav. Rev. (2015), http://dx.doi.org/10.1016/j.neubiorev.2015.01.011

G Model NBR-2117; No. of Pages 14 10

ARTICLE IN PRESS R.G. Mair et al. / Neuroscience and Biobehavioral Reviews xxx (2015) xxx–xxx

and Cavdar, 2013; Groenewegen, 1988; Jones, 2009; Rovó et al., 2012; Sherman and Guillery, 2002, 2011). The results of our PFC recording studies provide a basis for comparing coding properties of thalamic neurons and for examining the influence of thalamic inactivation of PFC function. To date, we have focused on analyzing the activity of neurons in MD and IL. We have recorded activity from 1122 well-isolated neurons in 5 rats (Miller et al., 2014). In general thalamic neurons have shown higher rates of spike firing and, perhaps as a result, less distinctive PETH patterns than PFC neurons. Fig. 6 compares rasters, PETHs, and place fields for movement and delay response types in MD and PFC. The thalamic histograms reveal similar temporal patterns of event related activity, seemingly imposed on top of a higher rate of background activity. Since our recordings are made from rats actively moving about the arena performing a challenging task, it is quite possible that higher firing rates in thalamus reflect the sensitivity of these neurons to behavioral arousal. Statistically significant PETH responses were observed for 268 (22.5%) of 1122 thalamic neurons, 159 (59&) of which fit the discrete categories observed for PFC. The finding of similar response patterns in medial thalamus and PFC is consistent with previous results for primates (Funahashi, 2013; Wyder et al., 2003) and rats (Han et al., 2013). Our results are consistent with previous reports that rat MD contains a preponderance of neurons responding to actions, action outcomes, and stimulus events and few if any neurons exhibiting prolonged delay-related responses comparable to the monkey (Han et al., 2013; Kawagoe et al., 2007; Oyoshi et al., 1996; Yu et al., 2012). The similarities in responses observed in MD, IL, and PFC seem consistent with evidence that PFC is an important source of driver/Class 1 inputs to these nuclei (Bay and Cavdar, 2013; Groenewegen, 1988; Rovó et al., 2012; Sherman and Guillery, 2002, 2011; Steriade et al., 1997) and the similarities between the behavioral effects of lesions damaging MD and PFC (reviewed above). Results to date indicate that thalamus has a much higher proportion of suppressive responses than PFC. This is presumably related to the prominent inhibitory inputs to MD and IL from ventral pallidum and substantia nigra and from the thalamic reticular nucleus (Groenewegen, 1988; Zikopoulos and Barbas, 2006). Other thalamic neurons exhibit elevated activity during the intertrial interval that is not time-locked to the end of the preceding trial or the start of the next trial. Presumably these reflect behavioral events when rats are not engaged in DNMTP responding. The relatively large number of thalamic responses that are distinct from medial PFC response types and the less well defined thalamic PETH responses are hard to reconcile with a role for MD as a primary driver of PFC responses, comparable to the influence of specific sensory nuclei such as the lateral geniculate nucleus on sensory cortical activity. Rather our results seem more consistent with the view that PFC responses are determined primarily by corticocortical inputs (Sherman and Guillery, 2002, 2011; Yeterian et al., 2012). The mix of MD and IL responses similar (59%) and distinct (41%) from medial PFC seems consistent with the mix of driver and modulatory inputs to these nuclei from medial PFC and from other areas of PFC, basal ganglia, limbic system, the thalamic reticular nucleus, and brain stem systems. 5.3. Effects of medial thalamic inactivation on prefrontal activity Sherman and Guillery (2002) have argued that “corticothalamocortical re-entry” pathways may provide higher order thalamic nuclei, such as MD and IL, with “an indispensable position for the modulation of messages involved in corticocortical processing.” To examine the potential of this influence we recorded the activity of medial PFC neurons while inhibiting medial thalamus with doses of musimol previously shown to impair DMTP activity when

microinjected bilaterally (Mair and Hembrook, 2008). Parnaudeau et al. (2013) used a designer receptor approach to produce mild inhibition of MD with systemic drug injections in mice while recording neurons in MD and PFC in awake, behaving mice. They found that inhibiting MD impaired performance on reversal learning and DMTP tasks and interfered with task-dependent modulation of beta-range synchrony between MD and PFC that was associated with acquisition and performance of the DMTP task. It is unclear from these results whether MD inhibition directly affected beta synchrony or if this effect was an indirect consequence of the behavioral impairment produced by this treatment. We examined the effects of inhibiting medial thalamus by microinjecting the GABAA agonist muscimol in central thalamus and recorded PFC activity in rats performing the DNMTP task. To avoid contaminating results by disrupting behavioral performance, we inactivated thalamus unilaterally and recorded activity in ipsilateral PFC. Consistent with previous results in our lab unilateral treatment had limited effects on behavioral performance (Porter et al., 2001). In inactivation studies (Wormwood et al., 2014) a guide cannula is aimed at central thalamus on one side of the brain while moveable tetrode arrays are implanted in ipsilateral PFC. The effects of reversible inactivation are analyzed by comparing recordings across three daily sessions in which the tetrode array is not moved (Fig. 7). Day 1 is without inactivation. This is typically a day when we find PFC neurons with robust signals and clear behavioral correlates. On day 2, thalamus is inactivated unilaterally with the GABAA agonist muscimol. On day 3, recordings are again carried out without inactivation. The identity of neurons across the three days of recording is established by comparing waveforms recorded by each of the microwire electrodes in the tetrode. Where we observe comparable PETH and place mapping results on day 1 and day 3 for a set of matched waveforms we compare results on day 2 to day 1 and day 3 to determine effects of inactivation (Fig. 7). To date, we have carried out successful inactivation studies with 40 PFC neurons comparing inactivation produced by doses from 0.25–2.0 nmol of muscimol in 0.5 ␮l of solution. Some PFC neurons exhibit increased and others decreased activity with thalamic inhibition. These effects did not seem related to the dose of musimol tested or the response type recorded. In fact both effects have been observed for neurons exhibiting the same response type during the same three day inactivation sequence. Our observation of mixed excitatory and inhibitory effects is consistent with optogenetic evidence that in mice core and matrix thalamocortical projections may have differential effects on excitatory and inhibitory mechanisms in prefrontal cortex (Cruikshank et al., 2012). Whether excited or inhibited prefrontal responses were consistently degraded by thalamic inhibition (Fig. 7). Purushothaman et al. (2012) carried similar inactivation studies examining the effects of another associative thalamic nucleus, the lateral pulvinar, and found an almost complete loss of visually evoked cortical responses in primary visual cortex, a result interpreted as evidence that pulvinar has a strong influence on cortical processing of sensory information. In testing a range of doses we found evidence of a powerful modulatory influence of medial thalamus on prefrontal activity. Comparable bilateral injections of musimol have been shown to impair DMTP performance (Mair and Hembrook, 2008) and similar impairment has been observed with bilateral infusions of lidocaine (Porter et al., 2001) and by lesions with localized damage in the area of the injection site (Burk and Mair, 1998; Bailey and Mair, 2005). We have not yet carried out anatomical control injections to test the localization of these treatments. Histological analyses of animals in these studies suggest that MD and IL are likely sites for these effects. Previous behavioral studies have identified MD and IL as likely sites of action for inactivation effects on DMTP based on anatomical controls (Mair and Hembrook, 2008), crossed inactivation analyses (Porter et al.,

Please cite this article in press as: Mair, R.G., et al., The neurobiology of thalamic amnesia: Contributions of medial thalamus and prefrontal cortex to delayed conditional discrimination. Neurosci. Biobehav. Rev. (2015), http://dx.doi.org/10.1016/j.neubiorev.2015.01.011

G Model NBR-2117; No. of Pages 14

ARTICLE IN PRESS R.G. Mair et al. / Neuroscience and Biobehavioral Reviews xxx (2015) xxx–xxx

Fig. 7. Effects of ipsilateral inactivation of medial thalamus with muscimol on a delay-related response in medial prefrontal cortex. Peri-event time histograms and rasters compare responses of a neuron with delay-related activity on three consecutive days: days 1 and 3 without inactivation and day 2 with inactivation from a microinjection of muscimol. Neurons were identified by waveforms recorded by the four microwire electrodes in the tetrode and cluster cutting analyses and these were confirmed by analyses of event-related activity and place field analyses on day 1 and day 3. These results illustrate typical effects of inactivation: reduced neural activity and a degraded event-related response that preserved the broad pattern of responses observed on days 1 and 3 when neurons were recorded without inactivation.

2001), and comparison with effects of discrete thalamic lesions (Bailey and Mair, 2005; Burk and Mair, 1998, 1999). 6. Medial thalamus, global amnesia, and executive function Global amnesia is characterized by an inability to recall and thus to respond based on recently presented declarative information in tasks used to assess anterograde memory with a relatively spared ability to recall well-established (context-free) declarative information for tasks such as vocabulary definition or general

11

information used to assess IQ. This is a distinction that is well established in cohort studies of WKS (McEntee et al., 1984; Butters and Cermak, 1980; Talland, 1965). Comparative neuropsychological studies of WKS and rats with medial thalamic lesions have described what seems a parallel distinction: impairment of delayed conditional discriminations, in which correct two alternative forced-choice responses are contingent on information remembered from the preceding sample response, with spared performance of well-established discriminations where similar choice responses are based on a fixed stimulus-response rule (Fig. 2). Other lesion studies have provided evidence that MD and PFC support more basic aspects of instrumental responding, such as action-outcome encoding and cue-guided response selection, that are integral to the ability to guide decision making flexibly as a function of environmental contingencies (Bradfield et al., 2013; Ostlund and Balleine, 2007, 2008). Electrophysiological recordings of medial thalamic and PFC neurons in awake behaving rats are consistent with these results. We found populations of neurons in thalamus and cortex with comparable event-related responses indicative of a critical role for MD and PFC in complex goal-directed behavior. Some neurons exhibited action-related activity associated with specific movement and lever press responses. Others responded in anticipation of reinforcement, following delivery of reinforcement, and signaling errors when expected reinforcement was not delivered. The preponderance of action- and reinforcement-related responses reveals a fundamental concern with action outcome relationships and cue-based action selection. Inactivation studies show that medial thalamus exerts critical control over the coding properties of PFC neurons. Inhibition of neurons centered on MD and IL substantially affects PFC activity and degrades signal-to-noise properties of event-related responses. Taken together, lesion and electrophysiological studies provide convergent evidence that medial thalamus and PFC support the ability to respond flexibly based on recent, current, or anticipated environmental circumstances. Saalmann (2014) has argued that midline, IL, and MD nuclei promote large-scale integration of information in cerebral cortex by modulating the degree of synchrony in distributed cortical circuits. Our results demonstrate that thalamic and PFC neurons exhibit precisely coordinated responses that encode diverse behavioral events in rats performing a flexible DNMTP task and they show that reversible inactivation of medial thalamus disrupts this activity in PFC. Our working hypothesis is that medial thalamus has an organizing or gating function, coordinating the activity of PFC with distributed cortical, limbic, and basal ganglia pathways that give rise to cognitive functions such as decision making, inhibitory control, supervisory attentional processes, and memory that are important for complex goal-directed behavior (León-Domínguez et al., 2013; Saalmann, 2014; Schiff et al., 2013).

References Aggleton, J.P., Brown, M.W., 1999. Episodic memory, amnesia and the hippocampalanterior thalamic axis. Behav. Brain Sci. 22, 425–444. Aggleton, J.P., Brown, M.W., 2006. Interleaving brain systems for episodic and recognition memory. Trends Cogn. Sci. 10, 455–463. Aggleton, J.P., Hunt, P.R., Nagle, S., Neave, N., 1996. The effects of selective lesions within the anterior thalamic nuclei on spatial memory in the rat. Behav. Brain Res. 81, 189–198. Aggleton, J.P., Nelson, A.D.J., 2014. Why do lesions in the rodent anterior thalamic nuclei cause such severe spatial deficits? Neurosci. Biobehav. Rev., http://dx.doi.org/10.1016/j.neubiorev.2014.08.013. Aggleton, J.P., Nicol, R.M., Huston, A.E., Fairbairn, A.F., 1988. The performance of amnesic subjects on tests of experimental amnesia in animals: delayed matching-to-sample and concurrent learning. Neuropsychologia 26, 265–272. Alexander, G.E., Fuster, J.M., 1973. Effects of cooling prefrontal cortex on cell firing in the nucleus medialis dorsalis. Brain Res. 61, 79–91. Alexander, W.H., Brown, J.W., 2011. Medial prefrontal cortex as an action-outcome predictor. Nat. Neurosci. 14, 1338–1344.

Please cite this article in press as: Mair, R.G., et al., The neurobiology of thalamic amnesia: Contributions of medial thalamus and prefrontal cortex to delayed conditional discrimination. Neurosci. Biobehav. Rev. (2015), http://dx.doi.org/10.1016/j.neubiorev.2015.01.011

G Model NBR-2117; No. of Pages 14 12

ARTICLE IN PRESS R.G. Mair et al. / Neuroscience and Biobehavioral Reviews xxx (2015) xxx–xxx

Bailey, K.R., Mair, R.G., 2004. Dissociable effects of frontal cortical lesions on measure of visuospatial attention and spatial working memory in the rat. Cereb. Cortex 14, 974–985. Bailey, K.R., Mair, R.G., 2005. Lesions of specific and nonspecific thalamic nuclei affect prefrontal cortex-dependent aspects of spatial working memory. Behav. Neurosci. 119, 410–419. Bailey, K.R., Mair, R.G., 2007. Effects of frontal cortex lesions on action sequence learning in the rat. Eur. J. Neurosci. 25, 2905–2915. Balleine, B.W., Dickinson, A., 1998. Goal-directed instrumental action: contingency and incentive learning and their cortical substrates. Neuropharmacology 37, 407–419. Barker, R.I.B., Warburton, E.C., 2013. Object-in-place associative recognition memory depends on glutamate receptor neurotransmission within two defined hippocampal-cortical circuits: a critical role for AMPA and NMDA receptors in the hippocampus, perirhinal, and prefrontal cortices. Cereb. Cortex, http://dx.doi.org/10.1093/cercor/bht245. Bay, H.H., Cavdar, S., 2013. Regional connections of the mediodorsal thalamic nucleus in the rat. J. Integr. Neurosci. 12, 201–219. Bayer, L., Eggermann, E., Saint-Mleux, B., Machard, D., Jones, B.E., Mühlethaler, M., Serafin, M., 2002. Selective action of orexin (hypocretin) on nonspecific thalamocortical projection neurons. J. Neurosci. 22, 7835–7839. Bradfield, L.A., Hart, G., Balleine, B.W., 2013. The role of the anterior, mediodorsal, and parafascicular thalamus in instrumental conditioning. Front. Syst. Neurosci. 7, http://dx.doi.org/10.3389/fnsys.2013.00051, Article 51. Burianova, H., Grady, C.L., 2007. Common and unique neural activation in autobiographical, episodic, and semantic retrieval. J. Cogn. Neurosci. 19, 1520–1534. Burk, J.A., Mair, R.G., 1998. Thalamic amnesia reconsidered: excitotoxic lesions of the intralaminar nuclei, but not the mediodorsal nucleus disrupt place DMTS performance in the rat (Rattus norvegicus). Behav. Neurosci. 112, 54–67. Burk, J.A., Mair, R.G., 1999. Delayed matching-to-sample trained with retractable levers is impaired by lesions of the intralaminar or ventromedial but not the laterodorsal thalamic nuclei. Psychobiology 27, 351–363. Burk, J.A., Mair, R.G., 2001. Effects of dorsal and ventral striatal lesions on delayed matching trained with retractable levers. Behav. Brain Res. 122, 67–78. Bussey, T.J., Muir, J.L., Everitt, B.J., Robbins, T.W., 1997. Triple dissociation of anterior cingulate, posterior cingulate, and medial frontal cortices on visual discrimination tasks using a touchscreen testing procedure for the rat. Behav. Neurosci. 111, 920–936. Butters, N., Cermak, L.S., 1980. Alcoholic Korsakoff Syndrome: An Information Processing Approach. Academic Press, NY. Carlesimo, G.A., Lombardi, M.G., Caltagirone, C., 2011. Vascular thalamic amnesia: a reappraisal. Neuropsychologia 49, 777–789. Chang, J.Y., Chen, L., Luo, F., Shi, L.H., Woodward, D.J., 2002. Neuronal responses in the frontal cortico-basal ganglia system during delayed matching-to-sample task: ensemble recording in freely moving rats. Exp. Brain Res. 142, 67–80. Chee, M.W.L., Choo, W.C., 2004. Functional imaging of working memory after 24 hr of total sleep deprivation. J. Neurosci. 24, 4560–4567. Chudasama, Y., 2011. Animal models of prefrontal executive function. Behav. Neurosci. 125, 327–343. Chudasama, Y., Bussey, T.J., Muir, J.L., 2001. Effects of selective thalamic and prelimbic cortex lesions on two types of visual discrimination and reversal learning. Eur. J. Neurosci. 14, 1009–1020. Corbit, L.H., Balleine, B.W., 2003. The role of prelimbic cortex in instrumental conditioning. Behav. Brain Res. 146, 145–157. Corbit, L.H., Muir, J.L., Balleine, B.W., 2003. Lesions of mediodorsal nucleus and anterior thalamic nuclei produce dissociable effects on instrumental conditioning in rats. Eur. J. Neurosci. 18, 1286–1294. Cowen, S.L., McNaughton, B.L., 2007. Selective delay activity in the medial prefrontal cortex of the rat: contribution of sensorimotor information and contingency. J. Neurophysiol. 98, 303–316. Cross, L., Brown, M.W., Aggleton, J.P., Warburton, E.C., 2012. The mediodorsal thalamic nucleus and the medial prefrontal cortex of the rat function together to support associative recognition and recency but not item recognition. Learn. Mem. 20, 41–50. Cruikshank, S.J., Ahmed, O.J., Stevens, T.R., Patrick, S.L., Gonzalez, A.N., Elmaleh, M., Connors, B.W., 2012. Thalamic control of layer 1 circuits in prefrontal cortex. J. Neurosci. 32, 17813–17823. Dalley, J.W., Cardinal, R.N., Robbins, T.W., 2004. Prefrontal executive and cognitive functions in rodents: neural and neurochemical substrates. Neurosci. Biobehav. Rev. 28, 771–784. Dumont, J.R., Aggleton, J.P., 2013. Dissociation of recognition and recency memory judgments after anterior thalamic nuclei lesions in rats. Behav. Neurosci. 127, 415–431. Euston, D.R., Gruber, A.J., McNaughton, B.L., 2012. The role of medial prefrontal cortex in memory and decision making. Neuron 76, 1057–1070. Fama, R., Pitel, A.L., Sullivan, E.V., 2012. Anterograde episodic memory in Korsakoff syndrome. Neuropsychol. Rev. 22, 92–104. Francoeur, M.J., Onos, K.D., Wormwood, B.A., Miller, R.L.A., Ellis, C.R., Chase, D.C., Cole, H.A., Gibson, B.M., Mair, R.G., 2014. Complex Encoding of Information about Actions and Context in Rat Prefrontal Cortex. Neuroscience Meeting Planner. Society for Neuroscience, Washington, DC (online). Funahashi, S., 2006. Prefrontal cortex and working memory processes. Neuroscience 139, 251–261. Funahashi, S., 2013. Thalamic mediodorsal nucleus and its participation in spatial working memory processes: comparison with the prefrontal cortex. Front. Syst. Neurosci. 7, http://dx.doi.org/10.3389/fnsys.2013.0036, Article 36.

Fuster, J.M., Alexander, G.E., 1971. Neuron activity related to short-term memory. Science 173, 652–654. Gallivan, J.P., McLean, D.A., Flanagan, J.R., Culham, J.C., 2013. Where one hand meets the other: limb-specific and action-dependent movement plans decoded from preparatory signals in single human frontoparietal brain areas. J. Neurosci. 33, 1991–2008. Gold, J.J., Squire, L.R., 2006. The anatomy of amnesia: neurohistological analyses of three new cases. Learn. Mem. 13, 699–710. Groenewegen, H.J., 1988. Organization of the afferent connections of the mediodorsal thalamic nucleus in the rat, related to the mediodorsal-prefrontal topography. Neuroscience 24, 379–431. Groenewegen, H.C., Berendse, H.W., 1994. The specificity of the ‘non-specific’ midline and intralaminar thalamic nuclei. Trends Neurosci. 17, 52–57. Haber, S.M., Calzavara, R., 2009. The cortico-basal ganglia integrative network: the role of thalamus. Brain Res. Bull. 78, 69–74. Han, J., Lee, J.H., Kim, M.J., Jung, M.W., 2013. Neural activity in mediodorsal nucleus of thalamus in rats performing a working memory task. Front. Neural Circuits 7, 128, http://dx.doi.org/10.3389/fncir.2013.00128. Harrison, L.M., Mair, R.G., 1996. A comparison of the effects of frontal cortical and thalamic lesions on measures of spatial learning and memory in the rat. Behav. Brain Res. 75, 195–206. Hazell, A.S., Butterworth, R.F., 2009. Update of cell damage mechanisms in thiamine deficiency: focus on oxidative stress, excitotoxicity, and inflammation. Alcohol Alcohol. 44, 141–147. Hazell, A.S., Rao, K.V., Danbolt, N.C., Pow, D.V., Butterworth, R.F., 2001. Selective down-regulation of the astrocyte glutamate transporters GLT-1 and GLAST within the medial thalamus in experimental Wernicke’s encephalopathy. J. Neurochem. 78, 560–568. Heidbreder, C.A., Groenewegen, H.J., 2003. The medial prefrontal cortex in the rat: evidence for a dorso-ventral distinction based upon functional and anatomical characteristics. Neurosci. Biobehav. Rev. 27, 555–579. Hembrook, J.R., Mair, R.G., 2011. Lesions of reuniens and rhomboid thalamic nuclei impair radial maze win-shift performance. Hippocampus 21, 815–826. Hembrook, J.R., Onos, K.D., Wormwood, B.A., Dennison, S.W., Cripps, A.C., Mair, R.G., 2011. Temporal Dynamics of Spatial Working Memory: Effects of Event-related Deep Brain Stimulation of Ventral Midline Thalamus in Rats. Neuroscience Meeting Planner. Society for Neuroscience, Washington, DC (online). Hoover, W.B., Vertes, R.P., 2007. Anatomical analysis of afferent projections to the medial prefrontal cortex in the rat. Brain Struct. Funct. 212, 149–179. Hoover, W.B., Vertes, R.P., 2012. Collateral projections from nucleus reuniens of thalamus to hippocampus and medial prefrontal cortex in the rat: a single and double retrograde fluorescent labeling study. Brain Struct. Funct. 217, 191–209. Horst, N.K., Laubach, M., 2012. Working with memory: evidence for a role for the medial prefrontal cortex in performance monitoring during spatial delayed alteration. J. Neurophysiol. 108, 3276–3288. Izquierdo, A., Murray, E.A., 2010. Functional interaction of medial mediodorsal thalamic nucleus but not nucleus accumbens with amygdala and orbital prefrontal cortex is essential for adaptive response selection after reinforcer devaluation. J. Neurosci. 30, 661–669. Izquierdo, A., Suda, R.K., Murray, E.A., 2004. Bilateral orbital prefrontal cortex lesions in rhesus monkeys disrupt choices guided by both reward value and reward contingency. J. Neurosci. 24, 7540–7548. Johnson, E.L., Knight, R.T., 2015. Intracranial recordings and human memory. Curr. Opin. Neurobiol. 31, 18–25. Jones, E.G., 2009. Synchrony in the interconnected circuitry of the thalamus and cerebral cortex. Ann. N. Y. Acad. Sci. 1157, 10–23. Jung, M.W., Qin, Y., McNaughton, B.L., Barnes, C.S., 1998. Firing characteristics of deep layer neurons in prefrontal cortex in rats performing spatial working memory tasks. Cereb. Cortex 8, 437–450. Kawagoe, T., Tamura, R., Uwano, T., Nishijo, H., Efuku, S., Ono, T., 2007. Neural correlates of stimulus-reward association in the rat mediodorsal nucleus. Neuroreport 18, 683–688. Kesner, R.P., Churchwell, J.C., 2011. An analysis of rat prefrontal cortex in mediating executive function. Neurobiol. Learn. Mem. 96, 417–431. Kessler, J., Markowitz, H.J., Otto, B., 1982. Subtle but distinct impairments of rats with chemical lesions in the thalamic mediodorsal nucleus, tested in a radial arm maze. J. Comp. Physiol. Psychol. 96, 712–720. Kinomura, S., Larsson, J., Gulyás, B., Roland, P.E., 1997. Activation by attention of the human reticular formation and thalamic intralaminar nuclei. Science 271, 512–515. Knoth, R.L., Mair, R.G., 1991. Rats recovered from pyrithiamine induced thiamine deficiency (PTD) treatment respond more slowly and less accurately in a pretrained non-matching to sample task. Behav. Neurosci. 105, 375–385. Koger, S.M., Mair, R.G., 1992. Depletion of cortical norepinephrine in rats by 6-OHDA does not impair performance on a pretrained delayed non-matching to sample task. Behav. Neurosci. 106, 718–721. Koger, S.M., Mair, R.G., 1994. A comparison of the effects of frontal cortical and thalamic lesions on measures of olfactory learning and memory in the rat. Behav. Neurosci. 108, 1088–1100. Krill, J.J., Harper, C.G., 2012. Neuroanatomy and neuropathology associated with Korsakoff’s syndrome. Neuropsychol. Rev. 22, 72–80. Krout, K.E., Belzer, R.E., Loewy, A.D., 2002. Brainstem projections to midline and intralaminar thalamic nuclei of the rat. J. Comp. Neurol. 448, 53–101. Langlais, P.J., Mair, R.G., 1990. Protective effects of the glutamate antagonist MK-801 on pyrithiamine induced lesions and amino acid changes in rat brain. J. Neurosci. 10, 1664–1674.

Please cite this article in press as: Mair, R.G., et al., The neurobiology of thalamic amnesia: Contributions of medial thalamus and prefrontal cortex to delayed conditional discrimination. Neurosci. Biobehav. Rev. (2015), http://dx.doi.org/10.1016/j.neubiorev.2015.01.011

G Model NBR-2117; No. of Pages 14

ARTICLE IN PRESS R.G. Mair et al. / Neuroscience and Biobehavioral Reviews xxx (2015) xxx–xxx

Langlais, P.J., Mair, R.G., Anderson, C.D., McEntee, W.J., 1987. Monoamines and metabolites in cortex and subcortical structures: normal distribution and the effects of thiamine deficiency in the rat. Brain Res. 421, 140–149. Langlais, P.J., Mair, R.G., Anderson, C.D., McEntee, W.J., 1988. Long-lasting changes in regional brain amino acids and monoamines in recovered pyrithiamine treated rats. Neurochem. Res. 13, 1199–1206. Langlais, P.J., Mandel, R.J., Mair, R.G., 1992. Diencephalic lesions, learning impairments, and intact retrograde memory following acute thiamine deficiency in the rat. Behav. Brain Res. 48, 177–185. Langlais, P.J., Zhang, S.X., 1993. Extracellular glutamate is increased in thalamus during thiamine deficiency-induced lesions and is blocked by MK-801. J. Neurochem. 61, 2175–2182. León-Domínguez, U., Vela-Bueno, A., Froufé-Torrex, M., León-Carrión, J., 2013. A chronometric functional sub-network in the thalamo-cortical system regulates the flow of neural information necessary for conscious cognitive processes. Neuropsychologia 51, 1336–1349. Mair, R.G., 1994. On the role of thalamic pathology in diencephalic amnesia. Rev. Neurosci. 5, 105–140. Mair, R.G., Anderson, C.D., Langlais, P.J., McEntee, W.J., 1985. Thiamine deficiency depletes cortical norepinephrine and impairs learning processes in the rat. Brain Res. 360, 273–284. Mair, R.G., Anderson, C.D., Langlais, P.J., McEntee, W.J., 1988. Behavioral impairments, brain lesions, and monoaminergic activity in the rat following recovery from a bout of thiamine deficiency. Behav. Brain Res. 27, 223–239. Mair, R.G., Burk, J.A., Porter, M.C., 1998. In the rat (Rattus norvegicus) lesions of frontal cortex, hippocampus, and the intralaminar thalamic nuclei have distinct effects on remembering. Behav. Neurosci. 112, 772–792. Mair, R.G., Burk, J.A., Porter, M.C., 2003. Impairment of radial maze delayed nonmatching after lesions of anterior thalamus and parahippocampal cortex. Behav. Neurosci. 117, 596–605. Mair, R.G., Ferguson, A.E., Knoth, R.L., Langlais, P.J., 1989. Widespread cortical denervation in rats following recovery from thiamine deficiency. Soc. Neurosci. Abstr. 15, 1104. Mair, R.G., Hembrook, J.R., 2008. Memory enhancement with event-related stimulation of the rostral intralaminar thalamic nuclei. J. Neurosci. 28, 14293–14300. Mair, R.G., Knoth, R., Rabchenuk, S., Langlais, P.J., 1991a. Impairment of olfactory, auditory, and spatial serial reversal learning in rats recovered from pyrithiamine induced thiamine deficiency (PTD) treatment. Behav. Neurosci. 105, 360–374. Mair, R.G., Koch, J.K., Newman, J.B., Howard, J.R., Burk, J.A., 2002. A double dissociation in striatum between serial reaction time and radial maze delayed nonmatching performance in rats. J. Neurosci. 22, 6756–6765. Mair, R.G., Lacourse, D.M., 1992. Radiofrequency lesions of thalamus produce delayed non-matching to sample impairments comparable to pyrithiamineinduced thiamine deficiency. Behav. Neurosci. 106, 634–645. Mair, R.G., Otto, T., Knoth, R., Rabchenuk, S., Langlais, P.J., 1991b. An analysis of aversively conditioned learning and memory in rats recovered from pyrithiamine induced thiamine deficiency (PTD) treatment. Behav. Neurosci. 105, 351–359. Mair, R.G., Robinson, J.K., Koger, S.M., Fox, G.D., Zhang, Y.P., 1992. Delayednonmatching-to-sample performance is impaired by extensive, but not by limited, lesions of the thalamic in the rat. Behav. Neurosci. 106, 646–656. Mair, W.P.G., Warrington, E.K., Weiskrantz, L., 1979. Memory disorders in Korsakoff’s psychosis: a neuropathological and neuropsychological investigation of two cases. Brain 102, 749–783. Malamud, N., Skillicorn, S.A., 1956. Relationship between the Wernicke and the Korsakoff syndrome. Arch. Neurol. Psychiatry 76, 585–596. Malkova, L., Gaffan, D., Murray, E.A., 1997. Excitotoxic lesions of the amygdala fail to produce impairment in visual learning for auditory secondary reinforcement but interfere with reinforce devaluation effects in rhesus monkeys. J. Neurosci. 17, 6011–6020. Matsumoto, N., Minamimoto, T., Graybiel, A.M., Kimura, M., 2001. Neurons in the thalamic CM–Pf complex supply striatal neurons with information about behaviorally significant sensory events. J. Neurophysiol. 85, 960–976. Mayes, A.R., Meudel, P.R., Mann, D., Pickering, A., 1988. Location of lesion in Korsakoff’s syndrome: neuropsychological and neuropathological data in two patients. Cortex 24, 367–388. McEntee, W.J., Biber, M.P., Perl, D.P., Benson, F.D., 1976. Diencephalic amnesia: a reappraisal. J. Neurol. Neurosurg. Psychiatry 39, 436–441. McEntee, W.J., Mair, R.G., Langlais, P.J., 1984. Neurochemical pathology in Korsakoff’s psychosis: implications for other cognitive disorders. Neurology 34, 648–652. Miller, R.L., Onos, K.D., Francoeur, M.J., Wormwood, B.A., Smedley, E.B., Theriault, C.J., Ryder, E.E., Gibson, B.M., Mair, R.G., 2014. Encoding of Information about Action and Context by Medial Thalamus in the Rat: Comparisons to Prefrontal Cortex. Neuroscience Meeting Planner. Society for Neuroscience, Washington, DC (online). Minamimoto, T., Kimura, M., 2002. Participation of the thalamic CM–Pf complex in attentional orienting. J. Neurophysiol. 87, 3090–3101. Mitchell, A.S., Browning, P.G.F., Baxter, M.G., 2007. Neurotoxic lesions of the medial mediodorsal nucleus of the thalamus disrupt reinforce devaluation effects in rhesus monkeys. J. Neurosci. 27, 11289–11295. Mitchell, A.S., Chakroborty, S., 2013. What does the mediodorsal thalamus do? Front. Syst. Neurosci. 7, 37, http://dx.doi.org/10.3389/fnsys.2013.00037. Mitchell, A.S., Dalyrmple-Alford, J.C., 2005. Dissociable memory effects after medial thalamus lesions in the rat. Eur. J. Neurosci. 22, 973–985. Mitchell, A.S., Dalyrmple-Alford, J.C., 2006. Lateral and anterior thalamic lesions impair independent memory systems. Learn. Mem. 13, 388–396.

13

Mumby, D.G., Cameli, L., Glenn, M.J., 1999. Impaired allocentric spatial working memory and intact retrograde memory after thalamic damage caused by thiamine deficiency in rats. Behav. Neurosci. 113, 42–50. Mumby, D.G., Mana, M.J., Pinel, J.P., David, E., Banks, K., 1995. Pyrithiamine-induced thiamine deficiency impairs object recognition in rats. Behav. Neurosci. 109, 1209–1214. Onos, K.D., Wormwood, B.A., Miller, R.L.A., Francoeur, M.J., Hebert, E.F., Blake, A.W., Gibson, B.M., Mair, R.G., 2013. Memory-Coding Properties of Prefrontal Cortical Neurons in the Rat. Neuroscience Meeting Planner. Society for Neuroscience, Washington, DC (online). Oscar-Berman, M., 2012. Function and dysfunction of prefrontal brain circuitry in alcoholic Korsakoff’s syndrome. Neuropsychol. Rev. 22, 154–169. Oscar-Berman, M., Bonner, R.T., 1989. Nonmatching- (oddity) and delayed nonmatching-to-sample performance in aging, alcoholic, and alcoholic Korsakoff individuals. Psychobiology 17, 424–430. Oscar-Berman, M., Hutner, N., Bonner, R.T., 1992. Visual and auditory spatial and nonspatial delayed-response performance by Korsakoff and non-Korsakoff alcoholic and aging individuals. Behav. Neurosci. 106, 613–622. Oscar-Berman, M., Zola-Morgan, S., 1980. Comparative neuropsychology and Korsakoff’s syndrome. I – Spatial and visual reversal learning. Neuropsychologia 18, 499–512. Ostlund, S.B., Balleine, B.W., 2005. Lesions of medial prefrontal cortex disrupt the acquisition but not the expression of goal-directed learning. J. Neurosci. 25, 7763–7770. Ostlund, S.B., Balleine, B.W., 2007. Orbitofrontal cortex mediates outcome encoding in Pavlovian but not instrumental conditioning. J. Neurosci. 27, 4819–4825. Ostlund, S.B., Balleine, B.W., 2008. Differential involvement of the basolateral amygdala and mediodorsal thalamus in instrumental action selection. J. Neurosci. 28, 4398–4405. Oyoshi, T., Nishijo, H., Asakiura, T., Takamura, Y., Ono, T., 1996. Emotional and behavioral correlates of mediodorsal thalamic neurons during associative learning in rats. J. Neurosci. 16, 5812–5829. Parnaudeau, S., O’Neill, P.-K., Bolkan, S.S., Ward, R.D., Abbas, A.I., Roth, B.L., Balsam, P.D., Gordon, J.A., Kellendonk, C., 2013. Inhibition if mediodorsal thalamus disrupts thalamofrontal connectivity and cognition. Neuron 77, 1151–1162. Paus, T., 2000. Functional anatomy of arousal and attention systems in the human brain. Prog. Brain Res. 126, 65–77. Pergola, G., Suchan, B., 2013. Associative learning beyond the medial temporal lobe: many actors on the memory stage. Front. Behav. Neurosci. 7, 162, http://dx.doi.org/10.3389/fnbeh.2013.00162. Pitel, A.-L., Chételat, G., Le Berre, A.P., Desgranges, B., Eustache, F., Beaunieux, H., 2012. Macrostructural abnormalities in Korsakoff syndrome compared with uncomplicated alcoholism. Neurology 78, 1330–1333. Porter, M.C., Burk, J.A., Mair, R.G., 2000. A comparison of the effects of hippocampal or prefrontal cortical lesions on three versions of delayed nonmatching-to-sample based on positional or spatial cues. Behav. Brain Res. 109, 69–81. Porter, M.C., Koch, J., Mair, R.G., 2001. Effects of reversible inactivation of thalamostriatal circuitry on delayed matching trained with retractable levers. Behav. Brain Res. 119, 61–69. Poucet, B., 1997. Searching for spatial unit firing in the prelimbic area of the rat medial prefrontal cortex. Behav. Brain Res. 84, 151–159. Pratt, W.E., Mizumori, S.J.Y., 2001. Neurons in rat medial prefrontal cortex show anticipatory rate changes to predictable differential rewards in a spatial memory task. Behav. Brain Res. 123, 165–183. Purushothaman, G., Marion, R., Li, K., Casagrande, V.A., 2012. Gating and control of primary visual cortex by pulvinar. Nat. Neurosci. 15, 905–912. Reed, L.J., Lasserson, D., Marsden, P., Stanhope, N., Stevens, T., Bello, F., Kingsley, D., Colchester, A., Kopelman, M.D., 2003. PDG-PET findings in the Wernicke–Korsakoff syndrome. Cortex 39, 1027–1045. Robinson, J.K., Mair, R.G., 1992. MK-801 prevents brain lesions and DNMTS deficits produced by pyrithiamine-induced encephalopathy in rats. Behav. Neurosci. 106, 623–633. Rovó, Z., Ulbert, I., Ascády, L., 2012. Drivers of the primate thalamus. J. Neurosci. 32, 17894–17908. Saalmann, Y.B., 2014. Intralaminar and medial thalamic influence on cortical synchrony, information transmission and cognition. Front. Syst. Neurosci. 8, 83, http://dx.doi.org/10.3389/fnsys.2014.00083. Savage, L.M., Hall, J.M., Resende, L.S., 2012. Translational rodent models of Korsakoff syndrome reveal the critical neuroanatomical substrates of memory dysfunction and recovery. Neuropsychol. Rev. 22, 195–209. Savage, L.M., Hall, J.M., Vetreno, R.P., 2011. Anterior thalamic lesions alter both hippocampal-dependent behavior and hippocampal acetylcholine release in the rat. Learn. Mem. 18, 751–758. Schiff, N.D., 2008. Central thalamic contributions to arousal regulation and neurological disorders of consciousness. Ann. N. Y. Acad. Sci. 1129, 105–118. Schiff, N.D., Shah, S.A., Hudson, A.E., Nauvel, T., Kalik, S.F., Purpura, K.P., 2013. Gating of attentional effort through the central thalamus. J. Neurophysiol. 109, 1152–1163. Schmanmann, J.D., 2003. Vascular syndromes of the thalamus. Stroke 34, 2264–2278. Shah, S.A., Schiff, N.D., 2012. Central thalamic deep brain stimulation for cognitive neuromodulation—a review of proposed mechanisms and investigational studies. Eur. J. Neurosci. 32, 1135–1144. Sherman, S.M., Guillery, R.W., 2002. The role of the thalamus in the flow of information to the cortex. Philos. Trans. R. Soc. Lond. B 357, 1695–1708.

Please cite this article in press as: Mair, R.G., et al., The neurobiology of thalamic amnesia: Contributions of medial thalamus and prefrontal cortex to delayed conditional discrimination. Neurosci. Biobehav. Rev. (2015), http://dx.doi.org/10.1016/j.neubiorev.2015.01.011

G Model NBR-2117; No. of Pages 14 14

ARTICLE IN PRESS R.G. Mair et al. / Neuroscience and Biobehavioral Reviews xxx (2015) xxx–xxx

Sherman, S.M., Guillery, R.W., 2011. Distinct functions for direct and transthalamic corticocortical connections. J. Neurophysiol. 106, 1068–1077. Smith, Y., Galvan, A., Ellender, T.J., Doig, N., Villalba, R.M., Huerta-Ocampo, I., Wichmann, R., Bolam, J.P., 2014. The thalamostriatal system in normal and diseased states. Front. Syst. Neurosci. 8, 5, http://dx.doi.org/10.3389/fnsys.2014.00005. Squire, L.R., Zola-Morgan, S., Chen, K.S., 1988. Human amnesia and animal models of amnesia: performance of amnesic patients on tests designed for the monkey. Behav. Neurosci. 102, 210–221. Squire, L.R., Amaral, D.G., Zola-Morgan, S., Kritchevsky, M., Press, G., 1989. Description of brain injury in the amnesic patient N.A. based on magnetic resonance imaging. Exp. Neurol. 105, 23–35. Staudigl, T., Zaehle, T., Voges, J., Hanslmayr, S., Esslinger, C., Hinrichs, R., Schmitt, F.C., Heinze, H.-J., Richardson-Klavehn, A., 2012. Memory signals from thalamus: early thalamocortical phase synchronization entrains gamma oscillations during long-term memory retrieval. Neuropsychologia 50, 3519–3527. Steriade, M., Jones, E.G., McCormick, D.A., 1997. Thalamus. Elsevier, Amsterdam. Stevens, A.A., Mair, R.G., 1998. Auditory conditional discrimination is impaired without an imposed delay in rats with lesions of either frontal cortex or medial thalamus. Psychobiology 26, 205–215. Strait, C.E., Blanchard, T.C., Hayden, B.Y., 2014. Reward value comparison via mutual inhibition in ventromedial prefrontal cortex. Neuron 18, 1357–1366. Sullivan, E.V., Pfefferbaum, A., 2009. Neuroimaging of the Wernicke–Korsakoff syndrome. Alcohol Alcohol. 44, 155–165. Sziklas, V., Petrides, M., 2007. Contribution of the anterior thalamic nuclei to conditional learning in rats. Hippocampus 17, 456–461. Talland, G.A., 1965. Deranged Memory. Academic Press, NY. Tanji, K., Suzuki, K., Fujii, T., Higano, S., Yamadori, A., 2003. A case of frontal network amnesia. J. Neurol. Neurosurg. Psychiatry 74, 106–109. Totah, N.K.B., Kim, Y.B., Homayoun, H., Moghaddam, B., 2009. Anterior cingulate neurons represent errors and preparatory attention within the same behavioral sequence. J. Neurosci. 29, 6418–6426. van der Werf, Y.D., Witter, M.P., Groenewegen, H.J., 2002. The intralaminar and midline nuclei of the thalamus. Anatomical and functional evidence for participation in processes of arousal and awareness. Brain Res. Rev. 39, 107–140. van der Werf, Y., Scheltens, P., Lindeboom, J., Witter, M.P., Uylings, H.B.M., Jolles, J., 2003. Deficits of memory, executive functioning and attention following infarction in the thalamus: a study of 22 cases with localized lesions. Neuropsychologia 41, 1330–1344. Vertes, R.P., Hoover, W.B., Rodriguez, J.J., 2012. Projections of the central nucleus of the thalamus in the rat: node in cortical, striatal and limbic forebrain circuitry. Neuroscience 219, 120–136. Victor, M., Adams, R.D., Collins, G.H., 1989. The Wernicke–Korsakoff Syndrome. F.A. Davis, Philadelphia.

von Cramon, D.Y., Hebel, N., Schuri, U., 1985. A contribution to the anatomical basis of thalamic amnesia. Brain 108, 993–1008. Watanabe, Y., Funahashi, S., 2012. Thalamic mediodorsal nucleus and working memory. Neurosci. Biobehav. Rev. 36, 134–142. Watrous, A.J., Fell, J., Ekstrom, A.D., Axmacher, N., 2015. More than spikes: common oscillatory mechanisms for content specific neural representations during perception and memory. Curr. Opin. Neurobiol. 31, 33–39. Williams, M., Pennybacker, J., 1954. Memory disturbances in third ventricle tumors. J. Neurol. Neurosurg. Psychiatry 17, 115–123. Wolff, M., Gibb, S.J., Dalrymple-Alford, J.C., 2006. Beyond spatial memory: the anterior thalamus and memory for temporal order of a sequence of odor cues. J. Neurosci. 26, 2907–2913. Wormwood, B.A., Francoeur, M.J., Onos, K.D., Miller, R.L.A., Lehet, C.R., Brasley, E.K., Manzo, L.E., Gibson, B.N., Mair, R.G., 2014. Encoding Information about Actions and Context in the Rat Prefrontal Cortex: Effects of Thalamic Inactivation. Neuroscience Meeting Planner. Society for Neuroscience, Washington, DC (online). Wyder, M.T., Massoglia, D.P., Stanford, T.R., 2003. Quantitative assessment of the timing and tuning of visual-related, saccade-related, and delay period activity in primate central thalamus. J. Neurophysiol. 90, 2029–2052. Yeterian, E.H., Pandya, D.N., Tomaiuolo, F., Petrides, M., 2012. The cortical connectivity of the prefrontal cortex in the monkey brain. Cortex 48, 58–81. Young, H.L., Stevens, A.A., Kivlahan, E.K., Mair, R.G., 1996. A comparison of temporal decay in place memory tasks in rats with lesions of thalamus, frontal cortex, and the hippocampal system. Behav. Neurosci. 110, 1244–1260. Yu, C., Fan, D., Lopez, A., Yin, H.H., 2012. Dynamic changes in single unit and gamma oscillations in a thalamocortical circuit during rapid instrumental learning. PLOS ONE 7, e50578. Zhang, S.X., Weilersbacher, G.S., Henderson, S.W., Corso, T., Olney, J.W., Langlais, P.J., 1995. Excitotoxic cytopathology, progression, and reversibility of thiamine deficiency-induced diencephalic lesions. J. Neuropathol. Exp. Neurol. 42, 255–267. Zhang, Y.P., Bailey, K.A., Toupin, M.M., Mair, R.G., 2005. Involvement of ventral pallidum in prefrontal cortex-dependent aspects of spatial working memory. Behav. Neurosci. 119, 399–409. Zhang, Y.P., Burk, J.A., Glode, B.M., Mair, R.G., 1998. The effects of thalamic and olfactory cortical lesions on continuous olfactory DNMTS and olfactory discrimination in the rat (Rattus norvegicus). Behav. Neurosci. 112, 39–53. Zikopoulos, B., Barbas, H., 2006. Prefrontal projections to the thalamic reticular nucleus form a unique circuit for attentional mechanisms. J. Neurosci. 26, 7348–7361. Zola-Morgan, S., Squire, L.R., 1985. Amnesia in monkeys following lesions of the mediodorsal nucleus of the thalamus. Ann. Neurol. 17, 558–564.

Please cite this article in press as: Mair, R.G., et al., The neurobiology of thalamic amnesia: Contributions of medial thalamus and prefrontal cortex to delayed conditional discrimination. Neurosci. Biobehav. Rev. (2015), http://dx.doi.org/10.1016/j.neubiorev.2015.01.011