- Email: [email protected]
Increased spreading activation in depression
Brain and Cognition 77 (2011) 265–270
Contents lists available at SciVerse ScienceDirect
Brain and Cognition journal homepage: www.elsevier.com/locate/b&c
Increased spreading activation in depression Paul S. Foster a,b,⇑, Raegan C. Yung a, Kaylei K. Branch a, Kristi Stringer c, Brad J. Ferguson d, William Sullivan a, Valeria Drago b,e a
Middle Tennessee State University, Murfreesboro, TN, United States University of Florida, Gainesville, FL, United States c University of Alabama at Birmingham, Birmingham, AL, United States d University of Missouri, Columbia, MO, United States e Laboratorio LENITEM, IRCCS San Giovanni di Dio Fatebenefratelli, Brescia, Italy b
a r t i c l e
i n f o
Article history: Accepted 1 August 2011 Available online 1 September 2011 Keywords: Spreading activation COWAT Animal Naming Word frequency Depression
a b s t r a c t The dopaminergic system is implicated in depressive disorders and research has also shown that dopamine constricts lexical/semantic networks by reducing spreading activation. Hence, depression, which is linked to reductions of dopamine, may be associated with increased spreading activation. However, research has generally found no effects of depression on spreading activation, using semantic priming paradigms. We used a different paradigm to investigate the relationship between depression and spreading activation, one based on word frequencies. Our sample included 97 undergraduates who completed the BDI-II and the Controlled Oral Word Association test as well as the Animal Naming test. The results indicated that the group scoring within the depressed ranged evidenced greater spreading activation as compared to those who scored within the normal range on the BDI-II. The implications of these results as they relate to creativity in depression is discussed. Ó 2011 Elsevier Inc. All rights reserved.
1. Introduction Collins and Loftus (1975) proposed a theory of spreading activation in which semantic memory nodes (e.g. cats) are organized into large networks consisting of concepts (e.g. animals). The nodes within the conceptual networks are highly interconnected through associative, bidirectional links such that activation of a node within a network is purported to spread along the associative links to related nodes or concepts. Further, the semantic nodes within a network (e.g. cats and dogs) are more strongly interconnected than semantic nodes from different conceptual networks (e.g. cats and screwdrivers). The strength of the connections between the semantic nodes within a conceptual network also varies, with some connections being stronger (e.g. cats and dogs) and other being weaker (e.g. cats and beavers). The strength of the connections between nodes is determined by production frequency norms, or the frequency of the use of the associations between nodes, which then determines the speed of the spreading activation. Hence, activation will spread more quickly through associated nodes that have been accessed more frequently, as compared to nodes that have been accessed or activated less frequently. The ⇑ Corresponding author. Address: Middle Tennessee State University, Psychology Department, 1500 Greenland Drive, Murfreesboro, TN 37132, United States. Fax: +1 615 898 5027. E-mail address: [email protected] (P.S. Foster). 0278-2626/$ - see front matter Ó 2011 Elsevier Inc. All rights reserved. doi:10.1016/j.bandc.2011.08.001
strength of connectivity is also likely related to the Hebbian principal that neurons (or the neuronal assemblies that comprise the semantic nodes) that fire together wire together. The extent or spread of activation is dependent on the strength of the initial activation of the node such that greater initial activation will result in greater spread of activation from that node. Activation of conceptual networks, however, decreases over time or with some intervening activity. Dopamine may represent the source by which these networks are activated or deactivated. Specifically, through the role of dopamine as both a neurotransmitter and neuromodulator (Cepeda & Levine, 1998), dopamine may have a prominent role in the modulation of activation of the semantic and conceptual networks. The dopaminergic system originates in the cells of the substantia nigra and the ventral tegmental area, which then send projections to the striatum, limbic system, and the frontal lobes (Afifi & Bergman, 1998). Hence, through these projections to the basal ganglia and cortex, dopamine may play an important role in modulating activation of the areas within the frontal lobes that control activation of semantic and lexical networks. The specific action may be through modulation of ‘the signal to noise ratio’. Specifically, Servan-Schreiber, Printz, and Cohen (1990) presented a model of the effects of catecholamines in a neural network such that catecholamines have the effect of improving the signal detection performance of the network, by dampening weak signals while simultaneously amplifying stronger signals.
266
P.S. Foster et al. / Brain and Cognition 77 (2011) 265–270
The role of dopamine as a neuromodulator in spreading activation has received support in numerous investigations. Kischka et al. (1996), by integrated the spreading activation theory of Collins and Loftus (1975) with the model of Servan-Schreiber et al. (1990), proposed that whereas a high signal-to-noise ratio would result in decreased spreading activation, a low signal-to-noise ratio would result in increased spreading activation. The results indicated that dopamine significantly reduced the indirect semantic priming effect but only marginally affected direct semantic priming. These results were interpreted as supporting the hypothesis that dopamine increases signal-to-noise ratio in semantic networks by reducing spreading semantic activation. Other investigations have supported these findings using similar paradigms (Angwin et al., 2004; Roesch-Ely et al., 2006). Additionally, patients with Parkinson’s disease (PD) have been found to possess increased spreading activation (Angwin, Coplan, Chenery, Murdoch, & Silburn, 2006), which is expected since PD is associated with a depletion of dopamine (Braak & Braak, 2000; Mink, 1996). Research on spreading activation has typically used a lexical decision task, or semantic priming paradigm, to examine how activation spreads through semantic and/or lexical networks. We have proposed and used a different paradigm based on the average word frequencies from the words generated on the Controlled Oral Word Association Test (COWAT). The COWAT is a measure of verbal fluency that requires the individual to generate as many words as possible, within 60 s, that begin with a specified letter (usually F, then A, then S). Research using lexical decision tasks has indicated that the reaction time for identifying high frequency words is significantly faster than for low frequency words (Allen, McNeal, & Kvak, 1992; Allen, Smith, Lien, Weber, & Madden, 1997; Allen, Wallace, & Weber, 1995). Hence, given the longer reaction time for low frequency words, it may be deduced that the adequate activation of the nodes that represent lower frequency words require greater spreading activation. Essentially, greater spreading activation is required to activate words that have lower frequencies, i.e. are further out in the semantic/lexical network. We conducted an investigation using this new paradigm to investigate spreading activation in patients with Parkinson’s disease (PD). Since PD is associated with reduced levels of dopamine, our hypothesis was that PD would be associated with increased spreading activation. The findings indicated that PD patients exhibited a significantly lower average word frequency than the controls, indicating greater spreading activation in the PD patients (Foster et al., 2008). Research has also associated the dopaminergic system with depression (Dailly, Chenu, Renard, & Bourin, 2004; Malhi, Parker, & Greenwood, 2005). Lower concentrations of metabolites of dopamine, such as homovanillic acid, have been found in the cerebrospinal fluid of depressed patients (Roy, de Jong, & Linnoila, 1989). As mentioned previously, the dopaminergic system projects to the frontal lobes and there also are greater concentrations of dopamine in the left hemisphere relative to the right hemisphere (De la Fuente-Fernandez, Kishore, Calne, Ruth, & Stoessl, 2000; Glick, Ross, & Hough, 1982). Repetitive transcranial magnetic stimulation (rTMS) is an effective treatment for depression (Martin et al., 2003) and research has indicated that rTMS applied over the left dorsolateral prefrontal region releases endogenous dopamine in patients with major depression (Pogarell et al., 2006). The left frontal lobe may also be involved in spreading activation. Specifically, relative to high frequency words, low frequency words generate significantly greater activation at the left frontal lobe (Carreiras, Mechelli, & Price, 2006; Halgren et al., 2002; Joubert et al., 2004). Taken together, the aforementioned research suggests that depression may be associated with enhanced spreading activation. However, research has indicated that depressed patients do not differ from normal, healthy controls in spreading activation. Specifically, no differences between depressed patients and controls
have been reported in research using a lexical decision task (Besche-Richard, Passerieux, & Hardy-Bayle, 2002; Georgieff, Dominey, Michel, Marie-cardine, & Dalery, 1998). Further, Dannlowski et al. (2006) used a word pronunciation task to measure spreading activation and also found no differences between depressed patients and normal controls. Given this discrepancy between the purported effects of dopamine on spreading activation in depressed individuals and the findings of the relevant research, we sought to investigate whether this relationship existed using our different paradigm for measuring spreading activation. Our hypothesis was that depressed individuals would be associated with significantly lower average word frequencies on the COWAT and the Animal Naming tests, i.e. greater spreading activation. Additionally, we sought to determine whether a relationship exists between the initial activation and the extent of the subsequent spreading activation. 2. Methods 2.1. Participants The participants included 97 (31 men and 66 women) undergraduate students from Middle Tennessee State University with an age range of 18–37 years (M = 20.90; SD = 3.52). To be considered for inclusion the participants had to exhibit a preference for the right hemibody as indicated by their score on the Coren Porac and Duncan Laterality Questionnaire (CPD). Specifically, the participants had to obtain a positive score on the CPD, although a total of six of the participants indicated a preference for using the left hand even though their total CPD score was positive. The participants had no history of significant head injury, psychological illness, or neurological diseases. All participants were treated in accordance with the ethical principles of the American Psychological Association and provided written informed consent. 2.2. Apparatus 2.2.1. Animal Naming (AN) The AN test requires the individuals to name as many different animals as possible within 60 s. The dependent variable consisted of the average word frequency for all animals named. 2.2.2. Beck depression inventory – II (BDI-II) The BDI-II (Beck, Steer, & Brown, 1996) is a 21 item self-report questionnaire used for measuring the severity of depression. The items of the BDI-II address problems related to numerous psychological, cognitive, and physiological symptoms. Each item is rated by the patient on a scale of 0–3, with a range of possible scores from 0 to 63. 2.2.3. Coren Porac and Duncan Laterality Questionnaire (CPD) The CPD (Coren, Porac, & Duncan, 1979) is a self-report questionnaire consisting of 13 questions assessing lateral preference for the hand, foot, eye, and ear. Responses are scored as +1 for ‘‘right’’, 1 for ‘‘left’’, and 0 for ‘‘both’’. Thus, the range of scores possible on the CPD are from –13 to +13. 2.2.4. Controlled Oral Word Association Test (COWAT) The COWAT is a measure of lexical fluency to confrontation requiring the individual to name as many words as possible within 60 s that begin with a specified letter (F, A, and S were used in this investigation). The participants were restricted from using proper nouns, numbers, and stem words with different endings. The primary dependent variable consisted of the average word frequency for the words generated on all three target letters.
267
P.S. Foster et al. / Brain and Cognition 77 (2011) 265–270
The study was approved by the Institutional Review Board of Middle Tennessee State University. Subsequent to providing written informed consent the AN, COWAT, BDI-II, and CPD were administered. The order of administration of the tests was randomized to control for potential carry-over effects. All tests were administered using the aforementioned standard procedures. The word frequency for each word generated on the COWAT and AN was then obtained. Specifically, the Francis–Kucera corpus (Francis & Kucera, 1982) was used to obtain the word frequencies. The word frequency for words that violated the administration rules of the COWAT and the AN test were not included in the analysis. For each participant, the average word frequency for all words generated on the COWAT and on the AN test was calculated and this average word frequency was used in conducting statistical analyses.
3. Results Groups of depressed versus not depressed participants were created by using the traditional cut-off score of 14 or higher on the BDI-II as indicative of at least mild depression. Our sample included 18 individuals whose BDI-II score was at or above the cutoff of 14 and 79 individuals with a score below this cut-off. Given this large difference in group sizes in regard to the number of individuals with a score in the depressed range we matched the subjects in sex and age. Matching the participants in sex and age also helped to control for these potential confounds. All of the men had a match with the exact same age, as was true for most of the women. The exception among the women included two participants who were nontraditional students in the Depressed group with ages of 26 and 32 years. The closest match among the NonDepressed group was found for these two women, with the ages of these controls being 25 and 29 years. In those cases where there were multiple exact matches available for sex and age we chose the participant with the lowest BDI-II score as the match. After the matching process was completed there were 18 participants in the Depressed group (5 men and 13 women) and 18 participants in the Non-Depressed group (5 men and 13 women). Following group assignment, the Non-Depressed group had an average BDI-II score of 2.83 (SD = 2.09) and the Depressed group had an average BDI-II score of 19.67 (SD = 4.95). The Non-Depressed and Depressed groups did not significantly differ in terms of their CPD score (Non-Depressed M = 9.44, SD = 2.92; Depressed M = 9.06, SD = 2.98). Initial analyses were conducted to determine whether the groups produced the same number of words on the COWAT and AN test, which would assist in determining whether the number of words represented a potential confound that needed to be controlled. The results of two separate oneway between groups ANOVAs indicated no significant difference, F(1, 34) = .23, p = .64, in the number of word produced on the COWAT between the Non-Depressed group (M = 34.56, SD = 10.79) and the Depressed group (M = 33.00, SD = 8.72). Also, no significant difference, F(1, 34) = .59, p = .45, was found in the number of words produced on the AN test between the Non-Depressed group (M = 16.61, SD = 3.73) and the Depressed group (M = 17.44, SD = 2.68). We next analyzed whether group differences existed in the average word frequencies from the COWAT and AN test. The results of a 2 (Sex: Men and Women) 2 (Depression: Non-Depressed and Depressed) between groups ANOVA, with the average word frequency from the COWAT as the dependent variable, indicated no significant main effect for Sex and no significant Sex Depression interaction. However, the main effect for Depression was significant, F(1, 32) = 4.37, p = .045, with the Depressed
group having a significantly lower average COWAT word frequency (M = 126.68, SD = 81.70) than the Non-Depressed group (M = 170.91, SD = 92.45). Using the average word frequency from the AN as the dependent variable, the same between groups ANOVA indicated no significant Sex or Depression main effect and no significant Sex Depression interaction. As a follow-up analysis we also conducted separate correlations between depression severity, as indexed by the BDI-II score, and average word frequency. The results indicated that for the Non-Depressed group there were no significant correlations between depression severity and the average COWAT word frequency (r = .18, p = .24) or the average AN word frequency (r = .21, p = .20). Likewise, for the Depressed group there were no significant correlations found between depression severity and the average COWAT word frequency (r = .27, p = .14) or the average AN word frequency (r = .29, p = .12). However, these correlations did not take into consideration the subjects who were excluded following the matching process. Hence, we also conducted these same correlations using the entire sample of 97 individuals. Once again, though, there were no significant correlations between depression severity and either the average COWAT word frequency (r = .12, p = .13) or the average AN word frequency (r = .10, p = .18). To analyze whether the strength of the initial activation is associated with the extent of subsequent spreading activation we obtained the word frequency for the first word produced from each letter of the COWAT and for the first animal named on the AN test. The average word frequency for the remaining words was then calculated, with an overall average consisting of all words from three letters of the COWAT being calculated. Correlations were then conducted between the initial word frequency and the average word frequency for the remaining words. The results indicated that no significant correlation existed for the Non-Depressed group for either the COWAT (r = .27, p = .14) or the AN test (r = .37, p = .07). However, a significant positive correlation was found for the Depressed group when examining the relationship between the COWAT initial word frequency and the average frequency of the remaining words (r = .54, p = .01, see Figs. 1 and 2 for scatterplots). The correlation for the AN test was not significant (r = .06, p = .41).
4. Discussion The results of our investigation supported the hypothesis that depression is associated with increased spreading activation. The 400 350
FAS Remaining Word Frequency
2.3. Procedure
300 250 200 150 100 50 0
0
50
100
150
200
250
300
350
400
450
FAS First Word Frequency Fig. 1. Relationship between first word frequency and remaining frequency for the Non-Depressed group.
268
P.S. Foster et al. / Brain and Cognition 77 (2011) 265–270
400
FAS Remaining Word Frequency
350 300 250 200 150 100 50 0
0
50
100
150
200
250
300
350
400
450
FAS First Word Frequency Fig. 2. Relationship between first word frequency and remaining frequency for the Depressed group.
reason for the discrepancy between our findings and the findings of previous research that has found no difference in spreading activation in depressed individuals may be the use of a different paradigm to measure spreading activation. Previous research has used a lexical decision task from within a semantic priming paradigm in which prime words are followed by target words that are presented to the participants. In contrast, our paradigm permits the individuals to generate their own words, and hence may represent more of a ‘‘free flow’’ of spreading activation that is not confined by words that are chosen for the individuals. Although the possibility exists that our present findings differ from those of previous investigations due to our use of a less restrictive method of measuring spreading activation, other possible explanations exist. Performance on verbal fluency tasks is known to be related to the integrity of left frontal lobe functioning (Henry & Crawford, 2004). Patients with lesions to the left frontal lobe perform worse on measures of verbal fluency, such as the COWAT, than patients with right hemisphere lesions (Baldo, Shimamura, Delis, Kramer, & Kaplan, 2001; Stuss et al., 1998). Investigations using functional imaging techniques have also implicated the left frontal lobe in verbal fluency (Elfgren & Risberg, 1998; Phelps, Hyder, Blamire, & Shulman, 1997; Wood, Saling, Abbott, & Jackson, 2001). Further, depression is known to be associated with left frontal lobe dysfunction. Strokes restricted to the left frontal lobe are associated with sadness and depression (Gainotti, 1972; Hama et al., 2007; Morris, Robinson, Raphael, & Hopwood, 1996; Shimoda & Robinson, 1999; Starkstein et al., 1989). Reduced cortical electrical activity has also been reported in studies using electroencephalography (EEG), with relative left frontal deactivation and right frontal activation in depressed individuals (Debener et al., 2000; Henriques & Davidson, 1991; Schaffer, Davidson, & Saron, 1983), including individuals with a history of depression (Henriques & Davidson, 1990; Vuga et al., 2006). Not surprisingly, research has found that depression is associated with worse performance on executive tests sensitive to frontal lobe functioning, including verbal fluency (Henry & Crawford, 2005), the Stroop Color-Word Test (Moritz, Birkner, & Kloss, 2002; Raskin, Friedman, & DiMascio, 1982), and the Wisconsin Card Sorting Test (Martin, Oren, & Boone, 1991). The possibility exists that the increased spreading activation in lexical networks associated with depression in the present study may have been partly due to altered search processes in lexical networks that arise from the exec-
utive dysfunction that characterizes this population. Specifically, the possibility exists that depressed individuals use different search strategies in selecting and retrieving words, strategies that are less organized or planned. Future research will need to be conducted to determine the veracity of this proposition. The results of the present investigation also provide support for the postulate that the strength of the initial activation determine the extent of the spreading activation, with greater initial activation generating greater spreading activation. However, this relationship was only found for the depressed group and using the data from the COWAT average word frequency. Previous research has not explored this component of Collins and Loftus (1975) theory of spreading activation, likely because lexical decision tasks are not suited for this purpose. The reason this relationship was only found for the depressed individuals may be due to the aforementioned research indicating left frontal lobe involvement in depression, verbal fluency, and the production of low frequency words. Also, the possibility exists that the relationship was only found for the COWAT average word frequencies due to a relatively restricted range with the word frequencies from the AN test. Specifically, the range of word frequencies is likely much greater with words that begin with a specified letter than with animals. When performing the AN test the individual is likely to use the names of common animals first and then with the remaining time generate names of much less common animals. However, on the COWAT there is generally a wider range of possibilities and hence word frequencies. As a result, word frequencies from the COWAT may represent a more sensitive index of spreading activation. The effect of depression on spreading activation found in the present investigation may have been due to reduced levels of dopamine in depression. However, the possibility exists that depression may also be associated with a lower level of cortical arousal. As mentioned previously, depression is associated with reduced cortical activity as evidenced by studies using EEG (Debener et al., 2000; Henriques & Davidson, 1991; Schaffer et al., 1983). A high level of cortical arousal narrows the associative field and hence may suppress the ability to make remote associations. Lowered arousal might allow these remote associations to emerge. Support for the postulate that the level of arousal might determine the size of neural networks comes from recent research by Contreras and Llinas (2001). Using high speed optical imaging, they electrically stimulated subcortical white matter in slices of guinea pig brain and recorded the area of activated neocortex. They found that with low frequency stimulation cortical activation is at first somewhat limited, but after few milliseconds this activation spreads to nearby areas. After high frequency stimulation, however, the cortical excitation remained fixed to a small column of neurons that were directly above the stimulating electrode. Intracellular recording from the neurons around this excited column during the rapid stimulation revealed increased inhibitory synaptic activity that probably inhibited the spread of activation to other areas. Regarding the relationship between spreading activation in lexical and semantic memory networks and cortical activity, research has supported the existence of this relationship. Specifically, research has found that ERP components are related to spreading activation, as indexed by performance on semantic priming tasks (Franklin, Dien, Neely, Huber, & Waterson, 2007; Wiese & Schweinberger, 2011). Additionally, we have previously reported a significant positive relationship between the ages of angry memories and changes in low and high beta EEG magnitude (Foster & Harrison, 2004). However, no research has used the paradigm of investigating spreading activation used in the present study in combination with measures of cortical activity. Additionally, research has not addressed how spreading activation and the resulting changes in cortical activity may be related to depression. Future research will need to determine whether such relationships exist.
P.S. Foster et al. / Brain and Cognition 77 (2011) 265–270
Some researchers have investigated the relationship between depression or depressive psychosis to creativity (Post, 1994, 1996). These investigators found that many of the most creative composers, scientists, artists and writers had depressive or bipolar disorders and that there is a high incidence of affective disorders in this population. A potential source of this enhanced creativity in depression may be the increased spreading activation found in this population. It has been hypothesized that alteration of the brain’s neurotransmitter systems, primarily a reduction of catecholamines including norepinephrine might increase creative thinking (McCarley, 1982). As mentioned earlier, Kischka et al. (1996) found increased spreading activation in normals after administration of levodopa. However, levodopa is a precursor of both dopamine and norepinephrine, and the administration of L-dopa to these individuals may have also increased the level of norepinephrine. Ghacibeh, Shenker, Shenal, Uthman, and Heilman (2006) studied the creativity of patients who had vagus nerve stimulation for medically intractable partial epilepsy. They hypothesized that since vagus nerve stimulation may activate the neurons in the locus coeruleus (LC), and this may result in increase release of brain norepinephrine, vagus nerve stimulation should reduce creativity and cognitive flexibility. Their findings were consistent with their hypothesis. In depression, there are also lower brain levels of norepinephrine, and many drugs used to treat depression elevate the brain’s levels of norepinephrine. Vagus nerve stimulation also alleviates depression (Milby, Halpern, & Baltuch, 2008). Hence, based on this literature it would be reasonable to assume that depression might facilitate creative innovation because it is associated with reduced levels of norepinephrine. Certainly, future research needs to be conducted to determine whether such relationships between depression, spreading activation, and creativity exist. Finally, we should note that we did not measure intellectual functioning in our subjects. Performance on priming tasks is associated with intellectual functioning (Komatsu, Naito, & Fuke, 1996; Mosley & Ells, 1994). Although, the potential impact of intellectual functioning affecting the present study is somewhat tempered by the fact that we used college students as our sample. The likelihood of individuals with borderline, or lower, intelligence being represented in our sample is rather small. However, that being said, it might still be important for future research to replicate the present findings while controlling for intellectual functioning.
References Afifi, A. K., & Bergman, R. A. (1998). Functional neuroanatomy: Text and atlas. New York: McGraw Hill. Allen, P. A., McNeal, M., & Kvak, D. (1992). Perhaps the lexicon is coded as a function of word frequency. Journal of Memory and Language, 31, 826–844. Allen, P. A., Smith, A. F., Lien, M. C., Weber, T. A., & Madden, D. J. (1997). Word frequency effects at brief exposure durations: Comment on Paap and Johansen. Journal of Experimental Psychology: Human Perception and Performance, 23, 1792–1797. Allen, P. A., Wallace, B., & Weber, T. A. (1995). Influence of case type, word frequency, and exposure duration on visual word recognition. Journal of Experimental Psychology: Human Perception and Performance, 21, 914–934. Angwin, A., Chenery, H. J., Copland, D. A., Arnott, W. L., Murdoch, B. E., & Silburn, P. A. (2004). Dopamine and semantic activation: An investigation of masked direct and indirect priming. Journal of the International Neuropsychological Society, 10, 15–25. Angwin, A., Coplan, D. A., Chenery, H. J., Murdoch, B. E., & Silburn, P. A. (2006). The influence of dopamine on semantic activation in Parkinson’s disease: Evidence from a multipriming task. Neuropsychology, 20, 299–306. Baldo, J. V., Shimamura, A. P., Delis, D. C., Kramer, J., & Kaplan, E. (2001). Verbal and design fluency in patients with frontal lobe lesions. Journal of the International Neuropsychological Society, 7, 586–596. Beck, A. T., Steer, R. A., & Brown, G. K. (1996). Beck depression inventory-II. San Antonio TX: Harcourt Assessment, Inc.. Besche-Richard, C., Passerieux, C., & Hardy-Bayle, M. C. (2002). Lexical decision tasks in depressive patients: Semantic priming before and after clinical improvement. European Psychiatry, 17, 69–74. Braak, H., & Braak, E. (2000). Pathoanatomy of Parkinson’s disease. Journal of Neurology, 247(Suppl. 2), II3–II10.
269
Carreiras, M., Mechelli, A., & Price, C. J. (2006). Effect of word and syllable frequency on activation during lexical decision and reading aloud. Human Brain Mapping, 27, 963–972. Cepeda, C., & Levine, M. S. (1998). Dopamine and n-methyl-d-asparate receptor interactions in the neostriatum. Developmental Neuroscience, 20, 1–18. Collins, A. M., & Loftus, E. F. (1975). A spreading-activation theory of semantic processing. Psychological Review, 82, 407–428. Contreras, D., & Llinas, R. (2001). Voltage-sensitive dye imaging of neocortical spatiotemporal dynamics to afferent activation frequency. Journal of Neuroscience, 21, 9403–9413. Coren, S. P., Porac, C., & Duncan, P. (1979). A behaviorally validated self-report inventory to assess 4 types of lateral preferences. Journal of Clinical Neuropsychology, 1, 55–64. Dailly, E., Chenu, F., Renard, C. E., & Bourin, M. (2004). Dopamine, depression and antidepressants. Fundamentals of Clinical Pharmacology, 18, 601–607. Dannlowski, U., Kersting, A., Arolt, V., Lalee-Mentzel, J., Donges, U. S., & Suslow, T. (2006). Unimpaired automatic processing of verbal information in the course of clinical depression. Depression and Anxiety, 23, 325–330. De la Fuente-Fernandez, R., Kishore, A., Calne, D. B., Ruth, T. J., & Stoessl, A. J. (2000). Nigrostriatal dopamine system and motor lateralization. Behavioural Brain Research, 112, 63–68. Debener, S., Beauducel, A., Nessler, D., Brocke, B., Heilemann, H., & Kayser, J. (2000). Is resting anterior EEG alpha asymmetry a trait marker for depression? Neuropsychobiology, 41, 31–37. Elfgren, C. I., & Risberg, J. (1998). Lateralized frontal blood flow increases during fluency tasks: Influence of cognitive strategy. Neuropsychologia, 36, 505–512. Foster, P. S., Drago, V., FitzGerald, D. B., Skoblar, B. M., Crucian, G. P., & Heilman, K. M. (2008). Spreading activation of lexical-semantic networks in Parkinson’s disease. Neuropsychologia, 46, 1908–1914. Foster, P. S., & Harrison, D. W. (2004). Cerebral correlates of varying ages of emotional memories. Cognitive and Behavioral Neurology, 17, 85–92. Francis, W. N., & Kucera, H. (1982). Frequency analysis of English usage: Lexicon and grammar. Boston, MA: Houghton Mifflin Company. Franklin, M. S., Dien, J., Neely, J. H., Huber, E., & Waterson, L. D. (2007). Semantic priming modulates the N400, N300, and N400RP. Clinical Neurophysiology, 118, 1053–1068. Gainotti, G. (1972). Emotional behavior and hemispheric side of the lesion. Cortex, 8, 41–55. Georgieff, N., Dominey, P. F., Michel, F., Marie-cardine, M., & Dalery, J. (1998). Semantic priming in major depressive state. Psychiatry Research, 78, 29–44. Ghacibeh, G. A., Shenker, J. I., Shenal, B., Uthman, B. M., & Heilman, K. M. (2006). Effect of vagus nerve stimulation on creativity and cognitive flexibility. Epilepsy & Behavior, 8, 720–725. Glick, S. D., Ross, D. A., & Hough, L. B. (1982). Lateral asymmetry of neurotransmitters in human brain. Brain Research, 18, 53–63. Halgren, E., Dhond, R. P., Christensen, N., Van Petten, C., Marinkovic, K., Lewine, J. D., et al. (2002). N400-like magnetoencephalography responses modulated by semantic context, word frequency, and lexical class in sentences. NeuroImage, 17, 1101–1116. Hama, S., Yamashita, H., Shigenobu, M., Watanabe, A., Kurisu, K., Yamawaki, Kitaoka., et al. (2007). Post-stroke affective or apathetic depression and lesion location: Left frontal lobe and bilateral basal ganglia. European Archives of Psychiatry and Clinical Neuroscience, 257, 149–152. Henriques, J. B., & Davidson, R. J. (1990). Regional brain electrical asymmetries discriminate between previously depressed and healthy control subjects. Journal of Abnormal Psychology, 99, 22–31. Henriques, J. B., & Davidson, R. J. (1991). Left frontal hypoactivation in depression. Journal of Abnormal Psychology, 100, 535–545. Henry, J. D., & Crawford, J. R. (2004). A meta-analytic review of verbal fluency performance following focal cortical lesions. Neuropsychology, 18, 284–295. Henry, J. D., & Crawford, J. R. (2005). A meta-analytic review of verbal fluency deficits in depression. Journal of Clinical and Experimental Neuropsychology, 27, 78–101. Joubert, S., Beauregard, M., Walter, N., Bourgouin, P., Beaudoin, G., Leroux, J. M., et al. (2004). Neural correlates of lexical and sublexical processes in reading. Brain and Language, 89, 9–20. Kischka, U., Kammer, T. H., Maier, S., Weisbrod, M., Thimm, M., & Spitzer, M. (1996). Dopaminergic modulation of semantic network activation. Neuropsychologia, 34, 1107–1113. Komatsu, S., Naito, M., & Fuke, T. (1996). Age-related and intelligence-related differences in implicit memory: Effects of generation on a word-fragment completion test. Journal of Experimental Child Psychology, 62, 151–172. Malhi, G. S., Parker, G. B., & Greenwood, J. (2005). Structural and functional models of depression: From sub-types to substrates. Acta Psychiatrica Scand., 111, 94–105. Martin, J. L., Barbanoj, M. J., Schlaepfer, T. E., Thompson, E., Perez, V., & Kulisevsky, J. (2003). Repetitive transcranial magnetic stimulation for the treatment of depression: Systematic review and meta-analysis. British Journal of Psychiatry, 182, 480–491. Martin, D. J., Oren, Z., & Boone, K. (1991). Major depressives’ and dysthymics’ performance on the Wisconsin Card Sorting Test. Journal of Clinical Psychology, 47, 685–690. McCarley, R. W. (1982). REM sleep and depression: Common neurobiological control mechanisms. American Journal of Psychiatry, 139, 565–570. Milby, A. H., Halpern, C. H., & Baltuch, G. H. (2008). Vagus nerve stimulation for epilepsy and depression. Neurotherapeutics, 5, 75–85.
270
P.S. Foster et al. / Brain and Cognition 77 (2011) 265–270
Mink, J. W. (1996). The basal ganglia: Focused selection and inhibition of competing motor programs. Progress in Neurobiology, 50, 381–425. Moritz, S., Birkner, C., & Kloss, M. (2002). Executive functioning in obsessive– compulsive disorder, unipolar depression, and schizophrenia. Archives of Clinical Neuropsychology, 17, 477–483. Morris, P. L., Robinson, R. G., Raphael, B., & Hopwood, M. J. (1996). Lesion location and poststroke depression. Journal of Neuropsychiatry and Clinical Neuroscience, 8, 399–403. Mosley, J. L., & Ells, J. G. (1994). Automatic processing in mildly retarded and nonretarded persons. Journal of Intellectual Disability Research, 38, 599–612. Phelps, E. A., Hyder, F., Blamire, A. M., & Shulman, R. G. (1997). FMRI of the prefrontal cortex during overt verbal fluency. NeuroReport, 8, 561–565. Pogarell, O., Koch, W., Popperl, G., Tatsch, K., Jakob, F., Zwanzger, P., et al. (2006). Striatal dopamine release after prefrontal repetitive transcranial magnetic stimulation in major depression: Preliminary results of a dynamic [123I] IBZM SPECT study. Journal of Psychiatric Research, 40, 307–314. Post, F. (1994). Creativity and psychopathology. A study of 291 world-famous men. British Journal of Psychiatry, 165, 22–34. Post, F. (1996). Verbal creativity, depression and alcoholism: An investigation of one hundred American and British writers. The British Journal of Psychiatry, 168, 545–555. Raskin, A., Friedman, A. S., & DiMascio, A. (1982). Cognitive and performance deficits in depression. Psychopharmacology Bulletin, 18, 196–202. Roesch-Ely, D., Weiland, S., Scheffel, H., Schwaninger, M., Hundemer, H. P., Kolter, T., et al. (2006). Dopaminergic modulation of semantic priming in healthy volunteers. Biological Psychiatry, 60, 604–611. Roy, A., de Jong, J., & Linnoila, M. (1989). Cerebrospinal fluid monoamine metabolites and suicidal behavior in depressed patients: A 5-year follow-up study. Archives of General Psychiatry, 46, 609–612.
Schaffer, C. E., Davidson, R. J., & Saron, C. (1983). Frontal and parietal electroencephalogram asymmetry in depressed and nondepressed subjects. Biological Psychiatry, 18, 753–762. Servan-Schreiber, D., Printz, H., & Cohen, J. D. (1990). A network model of catecholamine effects: Gain, signal-to-noise ratio, and behavior. Science, 49, 892–895. Shimoda, K., & Robinson, R. G. (1999). The relationship between poststroke depression and lesion location in long-term follow-up. Biological Psychiatry, 45, 187–192. Starkstein, S. E., Robinson, R. G., Honig, M. A., Parikh, R. M., Joselyn, J., & Price, T. R. (1989). Mood changes after right-hemisphere lesions. British Journal of Psychiatry, 155, 79–85. Stuss, D. T., Alexander, M. P., Hamer, L., Palumbo, C., Dempster, R., Binns, M., et al. (1998). The effects of focal anterior and posterior brain lesions on verbal fluency. Journal of the International Neuropsychological Society, 4, 265–278. Vuga, M., Fox, N. A., Cohn, J. F., George, C. J., Levenstein, R. M., & Kovacs, M. (2006). Long-term stability of frontal electroencephalographic asymmetry in adults with a history of depression and controls. International Journal of Psychophysiology, 59, 107–115. Wiese, H., & Schweinberger, S. R. (2011). Accessing semantic person knowledge: Temporal dynamics of nonstrategic categorical and associative priming. Journal of Cognitive Neuroscience, 23, 447–459. Wood, A. G., Saling, M. M., Abbott, D. F., & Jackson, G. D. (2001). A neurocognitive account of frontal lobe involvement in orthographic lexical retrieval: An fMRI study. NeuroImage, 14, 162–169.
Contents lists available at SciVerse ScienceDirect
Brain and Cognition journal homepage: www.elsevier.com/locate/b&c
Increased spreading activation in depression Paul S. Foster a,b,⇑, Raegan C. Yung a, Kaylei K. Branch a, Kristi Stringer c, Brad J. Ferguson d, William Sullivan a, Valeria Drago b,e a
Middle Tennessee State University, Murfreesboro, TN, United States University of Florida, Gainesville, FL, United States c University of Alabama at Birmingham, Birmingham, AL, United States d University of Missouri, Columbia, MO, United States e Laboratorio LENITEM, IRCCS San Giovanni di Dio Fatebenefratelli, Brescia, Italy b
a r t i c l e
i n f o
Article history: Accepted 1 August 2011 Available online 1 September 2011 Keywords: Spreading activation COWAT Animal Naming Word frequency Depression
a b s t r a c t The dopaminergic system is implicated in depressive disorders and research has also shown that dopamine constricts lexical/semantic networks by reducing spreading activation. Hence, depression, which is linked to reductions of dopamine, may be associated with increased spreading activation. However, research has generally found no effects of depression on spreading activation, using semantic priming paradigms. We used a different paradigm to investigate the relationship between depression and spreading activation, one based on word frequencies. Our sample included 97 undergraduates who completed the BDI-II and the Controlled Oral Word Association test as well as the Animal Naming test. The results indicated that the group scoring within the depressed ranged evidenced greater spreading activation as compared to those who scored within the normal range on the BDI-II. The implications of these results as they relate to creativity in depression is discussed. Ó 2011 Elsevier Inc. All rights reserved.
1. Introduction Collins and Loftus (1975) proposed a theory of spreading activation in which semantic memory nodes (e.g. cats) are organized into large networks consisting of concepts (e.g. animals). The nodes within the conceptual networks are highly interconnected through associative, bidirectional links such that activation of a node within a network is purported to spread along the associative links to related nodes or concepts. Further, the semantic nodes within a network (e.g. cats and dogs) are more strongly interconnected than semantic nodes from different conceptual networks (e.g. cats and screwdrivers). The strength of the connections between the semantic nodes within a conceptual network also varies, with some connections being stronger (e.g. cats and dogs) and other being weaker (e.g. cats and beavers). The strength of the connections between nodes is determined by production frequency norms, or the frequency of the use of the associations between nodes, which then determines the speed of the spreading activation. Hence, activation will spread more quickly through associated nodes that have been accessed more frequently, as compared to nodes that have been accessed or activated less frequently. The ⇑ Corresponding author. Address: Middle Tennessee State University, Psychology Department, 1500 Greenland Drive, Murfreesboro, TN 37132, United States. Fax: +1 615 898 5027. E-mail address: [email protected] (P.S. Foster). 0278-2626/$ - see front matter Ó 2011 Elsevier Inc. All rights reserved. doi:10.1016/j.bandc.2011.08.001
strength of connectivity is also likely related to the Hebbian principal that neurons (or the neuronal assemblies that comprise the semantic nodes) that fire together wire together. The extent or spread of activation is dependent on the strength of the initial activation of the node such that greater initial activation will result in greater spread of activation from that node. Activation of conceptual networks, however, decreases over time or with some intervening activity. Dopamine may represent the source by which these networks are activated or deactivated. Specifically, through the role of dopamine as both a neurotransmitter and neuromodulator (Cepeda & Levine, 1998), dopamine may have a prominent role in the modulation of activation of the semantic and conceptual networks. The dopaminergic system originates in the cells of the substantia nigra and the ventral tegmental area, which then send projections to the striatum, limbic system, and the frontal lobes (Afifi & Bergman, 1998). Hence, through these projections to the basal ganglia and cortex, dopamine may play an important role in modulating activation of the areas within the frontal lobes that control activation of semantic and lexical networks. The specific action may be through modulation of ‘the signal to noise ratio’. Specifically, Servan-Schreiber, Printz, and Cohen (1990) presented a model of the effects of catecholamines in a neural network such that catecholamines have the effect of improving the signal detection performance of the network, by dampening weak signals while simultaneously amplifying stronger signals.
266
P.S. Foster et al. / Brain and Cognition 77 (2011) 265–270
The role of dopamine as a neuromodulator in spreading activation has received support in numerous investigations. Kischka et al. (1996), by integrated the spreading activation theory of Collins and Loftus (1975) with the model of Servan-Schreiber et al. (1990), proposed that whereas a high signal-to-noise ratio would result in decreased spreading activation, a low signal-to-noise ratio would result in increased spreading activation. The results indicated that dopamine significantly reduced the indirect semantic priming effect but only marginally affected direct semantic priming. These results were interpreted as supporting the hypothesis that dopamine increases signal-to-noise ratio in semantic networks by reducing spreading semantic activation. Other investigations have supported these findings using similar paradigms (Angwin et al., 2004; Roesch-Ely et al., 2006). Additionally, patients with Parkinson’s disease (PD) have been found to possess increased spreading activation (Angwin, Coplan, Chenery, Murdoch, & Silburn, 2006), which is expected since PD is associated with a depletion of dopamine (Braak & Braak, 2000; Mink, 1996). Research on spreading activation has typically used a lexical decision task, or semantic priming paradigm, to examine how activation spreads through semantic and/or lexical networks. We have proposed and used a different paradigm based on the average word frequencies from the words generated on the Controlled Oral Word Association Test (COWAT). The COWAT is a measure of verbal fluency that requires the individual to generate as many words as possible, within 60 s, that begin with a specified letter (usually F, then A, then S). Research using lexical decision tasks has indicated that the reaction time for identifying high frequency words is significantly faster than for low frequency words (Allen, McNeal, & Kvak, 1992; Allen, Smith, Lien, Weber, & Madden, 1997; Allen, Wallace, & Weber, 1995). Hence, given the longer reaction time for low frequency words, it may be deduced that the adequate activation of the nodes that represent lower frequency words require greater spreading activation. Essentially, greater spreading activation is required to activate words that have lower frequencies, i.e. are further out in the semantic/lexical network. We conducted an investigation using this new paradigm to investigate spreading activation in patients with Parkinson’s disease (PD). Since PD is associated with reduced levels of dopamine, our hypothesis was that PD would be associated with increased spreading activation. The findings indicated that PD patients exhibited a significantly lower average word frequency than the controls, indicating greater spreading activation in the PD patients (Foster et al., 2008). Research has also associated the dopaminergic system with depression (Dailly, Chenu, Renard, & Bourin, 2004; Malhi, Parker, & Greenwood, 2005). Lower concentrations of metabolites of dopamine, such as homovanillic acid, have been found in the cerebrospinal fluid of depressed patients (Roy, de Jong, & Linnoila, 1989). As mentioned previously, the dopaminergic system projects to the frontal lobes and there also are greater concentrations of dopamine in the left hemisphere relative to the right hemisphere (De la Fuente-Fernandez, Kishore, Calne, Ruth, & Stoessl, 2000; Glick, Ross, & Hough, 1982). Repetitive transcranial magnetic stimulation (rTMS) is an effective treatment for depression (Martin et al., 2003) and research has indicated that rTMS applied over the left dorsolateral prefrontal region releases endogenous dopamine in patients with major depression (Pogarell et al., 2006). The left frontal lobe may also be involved in spreading activation. Specifically, relative to high frequency words, low frequency words generate significantly greater activation at the left frontal lobe (Carreiras, Mechelli, & Price, 2006; Halgren et al., 2002; Joubert et al., 2004). Taken together, the aforementioned research suggests that depression may be associated with enhanced spreading activation. However, research has indicated that depressed patients do not differ from normal, healthy controls in spreading activation. Specifically, no differences between depressed patients and controls
have been reported in research using a lexical decision task (Besche-Richard, Passerieux, & Hardy-Bayle, 2002; Georgieff, Dominey, Michel, Marie-cardine, & Dalery, 1998). Further, Dannlowski et al. (2006) used a word pronunciation task to measure spreading activation and also found no differences between depressed patients and normal controls. Given this discrepancy between the purported effects of dopamine on spreading activation in depressed individuals and the findings of the relevant research, we sought to investigate whether this relationship existed using our different paradigm for measuring spreading activation. Our hypothesis was that depressed individuals would be associated with significantly lower average word frequencies on the COWAT and the Animal Naming tests, i.e. greater spreading activation. Additionally, we sought to determine whether a relationship exists between the initial activation and the extent of the subsequent spreading activation. 2. Methods 2.1. Participants The participants included 97 (31 men and 66 women) undergraduate students from Middle Tennessee State University with an age range of 18–37 years (M = 20.90; SD = 3.52). To be considered for inclusion the participants had to exhibit a preference for the right hemibody as indicated by their score on the Coren Porac and Duncan Laterality Questionnaire (CPD). Specifically, the participants had to obtain a positive score on the CPD, although a total of six of the participants indicated a preference for using the left hand even though their total CPD score was positive. The participants had no history of significant head injury, psychological illness, or neurological diseases. All participants were treated in accordance with the ethical principles of the American Psychological Association and provided written informed consent. 2.2. Apparatus 2.2.1. Animal Naming (AN) The AN test requires the individuals to name as many different animals as possible within 60 s. The dependent variable consisted of the average word frequency for all animals named. 2.2.2. Beck depression inventory – II (BDI-II) The BDI-II (Beck, Steer, & Brown, 1996) is a 21 item self-report questionnaire used for measuring the severity of depression. The items of the BDI-II address problems related to numerous psychological, cognitive, and physiological symptoms. Each item is rated by the patient on a scale of 0–3, with a range of possible scores from 0 to 63. 2.2.3. Coren Porac and Duncan Laterality Questionnaire (CPD) The CPD (Coren, Porac, & Duncan, 1979) is a self-report questionnaire consisting of 13 questions assessing lateral preference for the hand, foot, eye, and ear. Responses are scored as +1 for ‘‘right’’, 1 for ‘‘left’’, and 0 for ‘‘both’’. Thus, the range of scores possible on the CPD are from –13 to +13. 2.2.4. Controlled Oral Word Association Test (COWAT) The COWAT is a measure of lexical fluency to confrontation requiring the individual to name as many words as possible within 60 s that begin with a specified letter (F, A, and S were used in this investigation). The participants were restricted from using proper nouns, numbers, and stem words with different endings. The primary dependent variable consisted of the average word frequency for the words generated on all three target letters.
267
P.S. Foster et al. / Brain and Cognition 77 (2011) 265–270
The study was approved by the Institutional Review Board of Middle Tennessee State University. Subsequent to providing written informed consent the AN, COWAT, BDI-II, and CPD were administered. The order of administration of the tests was randomized to control for potential carry-over effects. All tests were administered using the aforementioned standard procedures. The word frequency for each word generated on the COWAT and AN was then obtained. Specifically, the Francis–Kucera corpus (Francis & Kucera, 1982) was used to obtain the word frequencies. The word frequency for words that violated the administration rules of the COWAT and the AN test were not included in the analysis. For each participant, the average word frequency for all words generated on the COWAT and on the AN test was calculated and this average word frequency was used in conducting statistical analyses.
3. Results Groups of depressed versus not depressed participants were created by using the traditional cut-off score of 14 or higher on the BDI-II as indicative of at least mild depression. Our sample included 18 individuals whose BDI-II score was at or above the cutoff of 14 and 79 individuals with a score below this cut-off. Given this large difference in group sizes in regard to the number of individuals with a score in the depressed range we matched the subjects in sex and age. Matching the participants in sex and age also helped to control for these potential confounds. All of the men had a match with the exact same age, as was true for most of the women. The exception among the women included two participants who were nontraditional students in the Depressed group with ages of 26 and 32 years. The closest match among the NonDepressed group was found for these two women, with the ages of these controls being 25 and 29 years. In those cases where there were multiple exact matches available for sex and age we chose the participant with the lowest BDI-II score as the match. After the matching process was completed there were 18 participants in the Depressed group (5 men and 13 women) and 18 participants in the Non-Depressed group (5 men and 13 women). Following group assignment, the Non-Depressed group had an average BDI-II score of 2.83 (SD = 2.09) and the Depressed group had an average BDI-II score of 19.67 (SD = 4.95). The Non-Depressed and Depressed groups did not significantly differ in terms of their CPD score (Non-Depressed M = 9.44, SD = 2.92; Depressed M = 9.06, SD = 2.98). Initial analyses were conducted to determine whether the groups produced the same number of words on the COWAT and AN test, which would assist in determining whether the number of words represented a potential confound that needed to be controlled. The results of two separate oneway between groups ANOVAs indicated no significant difference, F(1, 34) = .23, p = .64, in the number of word produced on the COWAT between the Non-Depressed group (M = 34.56, SD = 10.79) and the Depressed group (M = 33.00, SD = 8.72). Also, no significant difference, F(1, 34) = .59, p = .45, was found in the number of words produced on the AN test between the Non-Depressed group (M = 16.61, SD = 3.73) and the Depressed group (M = 17.44, SD = 2.68). We next analyzed whether group differences existed in the average word frequencies from the COWAT and AN test. The results of a 2 (Sex: Men and Women) 2 (Depression: Non-Depressed and Depressed) between groups ANOVA, with the average word frequency from the COWAT as the dependent variable, indicated no significant main effect for Sex and no significant Sex Depression interaction. However, the main effect for Depression was significant, F(1, 32) = 4.37, p = .045, with the Depressed
group having a significantly lower average COWAT word frequency (M = 126.68, SD = 81.70) than the Non-Depressed group (M = 170.91, SD = 92.45). Using the average word frequency from the AN as the dependent variable, the same between groups ANOVA indicated no significant Sex or Depression main effect and no significant Sex Depression interaction. As a follow-up analysis we also conducted separate correlations between depression severity, as indexed by the BDI-II score, and average word frequency. The results indicated that for the Non-Depressed group there were no significant correlations between depression severity and the average COWAT word frequency (r = .18, p = .24) or the average AN word frequency (r = .21, p = .20). Likewise, for the Depressed group there were no significant correlations found between depression severity and the average COWAT word frequency (r = .27, p = .14) or the average AN word frequency (r = .29, p = .12). However, these correlations did not take into consideration the subjects who were excluded following the matching process. Hence, we also conducted these same correlations using the entire sample of 97 individuals. Once again, though, there were no significant correlations between depression severity and either the average COWAT word frequency (r = .12, p = .13) or the average AN word frequency (r = .10, p = .18). To analyze whether the strength of the initial activation is associated with the extent of subsequent spreading activation we obtained the word frequency for the first word produced from each letter of the COWAT and for the first animal named on the AN test. The average word frequency for the remaining words was then calculated, with an overall average consisting of all words from three letters of the COWAT being calculated. Correlations were then conducted between the initial word frequency and the average word frequency for the remaining words. The results indicated that no significant correlation existed for the Non-Depressed group for either the COWAT (r = .27, p = .14) or the AN test (r = .37, p = .07). However, a significant positive correlation was found for the Depressed group when examining the relationship between the COWAT initial word frequency and the average frequency of the remaining words (r = .54, p = .01, see Figs. 1 and 2 for scatterplots). The correlation for the AN test was not significant (r = .06, p = .41).
4. Discussion The results of our investigation supported the hypothesis that depression is associated with increased spreading activation. The 400 350
FAS Remaining Word Frequency
2.3. Procedure
300 250 200 150 100 50 0
0
50
100
150
200
250
300
350
400
450
FAS First Word Frequency Fig. 1. Relationship between first word frequency and remaining frequency for the Non-Depressed group.
268
P.S. Foster et al. / Brain and Cognition 77 (2011) 265–270
400
FAS Remaining Word Frequency
350 300 250 200 150 100 50 0
0
50
100
150
200
250
300
350
400
450
FAS First Word Frequency Fig. 2. Relationship between first word frequency and remaining frequency for the Depressed group.
reason for the discrepancy between our findings and the findings of previous research that has found no difference in spreading activation in depressed individuals may be the use of a different paradigm to measure spreading activation. Previous research has used a lexical decision task from within a semantic priming paradigm in which prime words are followed by target words that are presented to the participants. In contrast, our paradigm permits the individuals to generate their own words, and hence may represent more of a ‘‘free flow’’ of spreading activation that is not confined by words that are chosen for the individuals. Although the possibility exists that our present findings differ from those of previous investigations due to our use of a less restrictive method of measuring spreading activation, other possible explanations exist. Performance on verbal fluency tasks is known to be related to the integrity of left frontal lobe functioning (Henry & Crawford, 2004). Patients with lesions to the left frontal lobe perform worse on measures of verbal fluency, such as the COWAT, than patients with right hemisphere lesions (Baldo, Shimamura, Delis, Kramer, & Kaplan, 2001; Stuss et al., 1998). Investigations using functional imaging techniques have also implicated the left frontal lobe in verbal fluency (Elfgren & Risberg, 1998; Phelps, Hyder, Blamire, & Shulman, 1997; Wood, Saling, Abbott, & Jackson, 2001). Further, depression is known to be associated with left frontal lobe dysfunction. Strokes restricted to the left frontal lobe are associated with sadness and depression (Gainotti, 1972; Hama et al., 2007; Morris, Robinson, Raphael, & Hopwood, 1996; Shimoda & Robinson, 1999; Starkstein et al., 1989). Reduced cortical electrical activity has also been reported in studies using electroencephalography (EEG), with relative left frontal deactivation and right frontal activation in depressed individuals (Debener et al., 2000; Henriques & Davidson, 1991; Schaffer, Davidson, & Saron, 1983), including individuals with a history of depression (Henriques & Davidson, 1990; Vuga et al., 2006). Not surprisingly, research has found that depression is associated with worse performance on executive tests sensitive to frontal lobe functioning, including verbal fluency (Henry & Crawford, 2005), the Stroop Color-Word Test (Moritz, Birkner, & Kloss, 2002; Raskin, Friedman, & DiMascio, 1982), and the Wisconsin Card Sorting Test (Martin, Oren, & Boone, 1991). The possibility exists that the increased spreading activation in lexical networks associated with depression in the present study may have been partly due to altered search processes in lexical networks that arise from the exec-
utive dysfunction that characterizes this population. Specifically, the possibility exists that depressed individuals use different search strategies in selecting and retrieving words, strategies that are less organized or planned. Future research will need to be conducted to determine the veracity of this proposition. The results of the present investigation also provide support for the postulate that the strength of the initial activation determine the extent of the spreading activation, with greater initial activation generating greater spreading activation. However, this relationship was only found for the depressed group and using the data from the COWAT average word frequency. Previous research has not explored this component of Collins and Loftus (1975) theory of spreading activation, likely because lexical decision tasks are not suited for this purpose. The reason this relationship was only found for the depressed individuals may be due to the aforementioned research indicating left frontal lobe involvement in depression, verbal fluency, and the production of low frequency words. Also, the possibility exists that the relationship was only found for the COWAT average word frequencies due to a relatively restricted range with the word frequencies from the AN test. Specifically, the range of word frequencies is likely much greater with words that begin with a specified letter than with animals. When performing the AN test the individual is likely to use the names of common animals first and then with the remaining time generate names of much less common animals. However, on the COWAT there is generally a wider range of possibilities and hence word frequencies. As a result, word frequencies from the COWAT may represent a more sensitive index of spreading activation. The effect of depression on spreading activation found in the present investigation may have been due to reduced levels of dopamine in depression. However, the possibility exists that depression may also be associated with a lower level of cortical arousal. As mentioned previously, depression is associated with reduced cortical activity as evidenced by studies using EEG (Debener et al., 2000; Henriques & Davidson, 1991; Schaffer et al., 1983). A high level of cortical arousal narrows the associative field and hence may suppress the ability to make remote associations. Lowered arousal might allow these remote associations to emerge. Support for the postulate that the level of arousal might determine the size of neural networks comes from recent research by Contreras and Llinas (2001). Using high speed optical imaging, they electrically stimulated subcortical white matter in slices of guinea pig brain and recorded the area of activated neocortex. They found that with low frequency stimulation cortical activation is at first somewhat limited, but after few milliseconds this activation spreads to nearby areas. After high frequency stimulation, however, the cortical excitation remained fixed to a small column of neurons that were directly above the stimulating electrode. Intracellular recording from the neurons around this excited column during the rapid stimulation revealed increased inhibitory synaptic activity that probably inhibited the spread of activation to other areas. Regarding the relationship between spreading activation in lexical and semantic memory networks and cortical activity, research has supported the existence of this relationship. Specifically, research has found that ERP components are related to spreading activation, as indexed by performance on semantic priming tasks (Franklin, Dien, Neely, Huber, & Waterson, 2007; Wiese & Schweinberger, 2011). Additionally, we have previously reported a significant positive relationship between the ages of angry memories and changes in low and high beta EEG magnitude (Foster & Harrison, 2004). However, no research has used the paradigm of investigating spreading activation used in the present study in combination with measures of cortical activity. Additionally, research has not addressed how spreading activation and the resulting changes in cortical activity may be related to depression. Future research will need to determine whether such relationships exist.
P.S. Foster et al. / Brain and Cognition 77 (2011) 265–270
Some researchers have investigated the relationship between depression or depressive psychosis to creativity (Post, 1994, 1996). These investigators found that many of the most creative composers, scientists, artists and writers had depressive or bipolar disorders and that there is a high incidence of affective disorders in this population. A potential source of this enhanced creativity in depression may be the increased spreading activation found in this population. It has been hypothesized that alteration of the brain’s neurotransmitter systems, primarily a reduction of catecholamines including norepinephrine might increase creative thinking (McCarley, 1982). As mentioned earlier, Kischka et al. (1996) found increased spreading activation in normals after administration of levodopa. However, levodopa is a precursor of both dopamine and norepinephrine, and the administration of L-dopa to these individuals may have also increased the level of norepinephrine. Ghacibeh, Shenker, Shenal, Uthman, and Heilman (2006) studied the creativity of patients who had vagus nerve stimulation for medically intractable partial epilepsy. They hypothesized that since vagus nerve stimulation may activate the neurons in the locus coeruleus (LC), and this may result in increase release of brain norepinephrine, vagus nerve stimulation should reduce creativity and cognitive flexibility. Their findings were consistent with their hypothesis. In depression, there are also lower brain levels of norepinephrine, and many drugs used to treat depression elevate the brain’s levels of norepinephrine. Vagus nerve stimulation also alleviates depression (Milby, Halpern, & Baltuch, 2008). Hence, based on this literature it would be reasonable to assume that depression might facilitate creative innovation because it is associated with reduced levels of norepinephrine. Certainly, future research needs to be conducted to determine whether such relationships between depression, spreading activation, and creativity exist. Finally, we should note that we did not measure intellectual functioning in our subjects. Performance on priming tasks is associated with intellectual functioning (Komatsu, Naito, & Fuke, 1996; Mosley & Ells, 1994). Although, the potential impact of intellectual functioning affecting the present study is somewhat tempered by the fact that we used college students as our sample. The likelihood of individuals with borderline, or lower, intelligence being represented in our sample is rather small. However, that being said, it might still be important for future research to replicate the present findings while controlling for intellectual functioning.
References Afifi, A. K., & Bergman, R. A. (1998). Functional neuroanatomy: Text and atlas. New York: McGraw Hill. Allen, P. A., McNeal, M., & Kvak, D. (1992). Perhaps the lexicon is coded as a function of word frequency. Journal of Memory and Language, 31, 826–844. Allen, P. A., Smith, A. F., Lien, M. C., Weber, T. A., & Madden, D. J. (1997). Word frequency effects at brief exposure durations: Comment on Paap and Johansen. Journal of Experimental Psychology: Human Perception and Performance, 23, 1792–1797. Allen, P. A., Wallace, B., & Weber, T. A. (1995). Influence of case type, word frequency, and exposure duration on visual word recognition. Journal of Experimental Psychology: Human Perception and Performance, 21, 914–934. Angwin, A., Chenery, H. J., Copland, D. A., Arnott, W. L., Murdoch, B. E., & Silburn, P. A. (2004). Dopamine and semantic activation: An investigation of masked direct and indirect priming. Journal of the International Neuropsychological Society, 10, 15–25. Angwin, A., Coplan, D. A., Chenery, H. J., Murdoch, B. E., & Silburn, P. A. (2006). The influence of dopamine on semantic activation in Parkinson’s disease: Evidence from a multipriming task. Neuropsychology, 20, 299–306. Baldo, J. V., Shimamura, A. P., Delis, D. C., Kramer, J., & Kaplan, E. (2001). Verbal and design fluency in patients with frontal lobe lesions. Journal of the International Neuropsychological Society, 7, 586–596. Beck, A. T., Steer, R. A., & Brown, G. K. (1996). Beck depression inventory-II. San Antonio TX: Harcourt Assessment, Inc.. Besche-Richard, C., Passerieux, C., & Hardy-Bayle, M. C. (2002). Lexical decision tasks in depressive patients: Semantic priming before and after clinical improvement. European Psychiatry, 17, 69–74. Braak, H., & Braak, E. (2000). Pathoanatomy of Parkinson’s disease. Journal of Neurology, 247(Suppl. 2), II3–II10.
269
Carreiras, M., Mechelli, A., & Price, C. J. (2006). Effect of word and syllable frequency on activation during lexical decision and reading aloud. Human Brain Mapping, 27, 963–972. Cepeda, C., & Levine, M. S. (1998). Dopamine and n-methyl-d-asparate receptor interactions in the neostriatum. Developmental Neuroscience, 20, 1–18. Collins, A. M., & Loftus, E. F. (1975). A spreading-activation theory of semantic processing. Psychological Review, 82, 407–428. Contreras, D., & Llinas, R. (2001). Voltage-sensitive dye imaging of neocortical spatiotemporal dynamics to afferent activation frequency. Journal of Neuroscience, 21, 9403–9413. Coren, S. P., Porac, C., & Duncan, P. (1979). A behaviorally validated self-report inventory to assess 4 types of lateral preferences. Journal of Clinical Neuropsychology, 1, 55–64. Dailly, E., Chenu, F., Renard, C. E., & Bourin, M. (2004). Dopamine, depression and antidepressants. Fundamentals of Clinical Pharmacology, 18, 601–607. Dannlowski, U., Kersting, A., Arolt, V., Lalee-Mentzel, J., Donges, U. S., & Suslow, T. (2006). Unimpaired automatic processing of verbal information in the course of clinical depression. Depression and Anxiety, 23, 325–330. De la Fuente-Fernandez, R., Kishore, A., Calne, D. B., Ruth, T. J., & Stoessl, A. J. (2000). Nigrostriatal dopamine system and motor lateralization. Behavioural Brain Research, 112, 63–68. Debener, S., Beauducel, A., Nessler, D., Brocke, B., Heilemann, H., & Kayser, J. (2000). Is resting anterior EEG alpha asymmetry a trait marker for depression? Neuropsychobiology, 41, 31–37. Elfgren, C. I., & Risberg, J. (1998). Lateralized frontal blood flow increases during fluency tasks: Influence of cognitive strategy. Neuropsychologia, 36, 505–512. Foster, P. S., Drago, V., FitzGerald, D. B., Skoblar, B. M., Crucian, G. P., & Heilman, K. M. (2008). Spreading activation of lexical-semantic networks in Parkinson’s disease. Neuropsychologia, 46, 1908–1914. Foster, P. S., & Harrison, D. W. (2004). Cerebral correlates of varying ages of emotional memories. Cognitive and Behavioral Neurology, 17, 85–92. Francis, W. N., & Kucera, H. (1982). Frequency analysis of English usage: Lexicon and grammar. Boston, MA: Houghton Mifflin Company. Franklin, M. S., Dien, J., Neely, J. H., Huber, E., & Waterson, L. D. (2007). Semantic priming modulates the N400, N300, and N400RP. Clinical Neurophysiology, 118, 1053–1068. Gainotti, G. (1972). Emotional behavior and hemispheric side of the lesion. Cortex, 8, 41–55. Georgieff, N., Dominey, P. F., Michel, F., Marie-cardine, M., & Dalery, J. (1998). Semantic priming in major depressive state. Psychiatry Research, 78, 29–44. Ghacibeh, G. A., Shenker, J. I., Shenal, B., Uthman, B. M., & Heilman, K. M. (2006). Effect of vagus nerve stimulation on creativity and cognitive flexibility. Epilepsy & Behavior, 8, 720–725. Glick, S. D., Ross, D. A., & Hough, L. B. (1982). Lateral asymmetry of neurotransmitters in human brain. Brain Research, 18, 53–63. Halgren, E., Dhond, R. P., Christensen, N., Van Petten, C., Marinkovic, K., Lewine, J. D., et al. (2002). N400-like magnetoencephalography responses modulated by semantic context, word frequency, and lexical class in sentences. NeuroImage, 17, 1101–1116. Hama, S., Yamashita, H., Shigenobu, M., Watanabe, A., Kurisu, K., Yamawaki, Kitaoka., et al. (2007). Post-stroke affective or apathetic depression and lesion location: Left frontal lobe and bilateral basal ganglia. European Archives of Psychiatry and Clinical Neuroscience, 257, 149–152. Henriques, J. B., & Davidson, R. J. (1990). Regional brain electrical asymmetries discriminate between previously depressed and healthy control subjects. Journal of Abnormal Psychology, 99, 22–31. Henriques, J. B., & Davidson, R. J. (1991). Left frontal hypoactivation in depression. Journal of Abnormal Psychology, 100, 535–545. Henry, J. D., & Crawford, J. R. (2004). A meta-analytic review of verbal fluency performance following focal cortical lesions. Neuropsychology, 18, 284–295. Henry, J. D., & Crawford, J. R. (2005). A meta-analytic review of verbal fluency deficits in depression. Journal of Clinical and Experimental Neuropsychology, 27, 78–101. Joubert, S., Beauregard, M., Walter, N., Bourgouin, P., Beaudoin, G., Leroux, J. M., et al. (2004). Neural correlates of lexical and sublexical processes in reading. Brain and Language, 89, 9–20. Kischka, U., Kammer, T. H., Maier, S., Weisbrod, M., Thimm, M., & Spitzer, M. (1996). Dopaminergic modulation of semantic network activation. Neuropsychologia, 34, 1107–1113. Komatsu, S., Naito, M., & Fuke, T. (1996). Age-related and intelligence-related differences in implicit memory: Effects of generation on a word-fragment completion test. Journal of Experimental Child Psychology, 62, 151–172. Malhi, G. S., Parker, G. B., & Greenwood, J. (2005). Structural and functional models of depression: From sub-types to substrates. Acta Psychiatrica Scand., 111, 94–105. Martin, J. L., Barbanoj, M. J., Schlaepfer, T. E., Thompson, E., Perez, V., & Kulisevsky, J. (2003). Repetitive transcranial magnetic stimulation for the treatment of depression: Systematic review and meta-analysis. British Journal of Psychiatry, 182, 480–491. Martin, D. J., Oren, Z., & Boone, K. (1991). Major depressives’ and dysthymics’ performance on the Wisconsin Card Sorting Test. Journal of Clinical Psychology, 47, 685–690. McCarley, R. W. (1982). REM sleep and depression: Common neurobiological control mechanisms. American Journal of Psychiatry, 139, 565–570. Milby, A. H., Halpern, C. H., & Baltuch, G. H. (2008). Vagus nerve stimulation for epilepsy and depression. Neurotherapeutics, 5, 75–85.
270
P.S. Foster et al. / Brain and Cognition 77 (2011) 265–270
Mink, J. W. (1996). The basal ganglia: Focused selection and inhibition of competing motor programs. Progress in Neurobiology, 50, 381–425. Moritz, S., Birkner, C., & Kloss, M. (2002). Executive functioning in obsessive– compulsive disorder, unipolar depression, and schizophrenia. Archives of Clinical Neuropsychology, 17, 477–483. Morris, P. L., Robinson, R. G., Raphael, B., & Hopwood, M. J. (1996). Lesion location and poststroke depression. Journal of Neuropsychiatry and Clinical Neuroscience, 8, 399–403. Mosley, J. L., & Ells, J. G. (1994). Automatic processing in mildly retarded and nonretarded persons. Journal of Intellectual Disability Research, 38, 599–612. Phelps, E. A., Hyder, F., Blamire, A. M., & Shulman, R. G. (1997). FMRI of the prefrontal cortex during overt verbal fluency. NeuroReport, 8, 561–565. Pogarell, O., Koch, W., Popperl, G., Tatsch, K., Jakob, F., Zwanzger, P., et al. (2006). Striatal dopamine release after prefrontal repetitive transcranial magnetic stimulation in major depression: Preliminary results of a dynamic [123I] IBZM SPECT study. Journal of Psychiatric Research, 40, 307–314. Post, F. (1994). Creativity and psychopathology. A study of 291 world-famous men. British Journal of Psychiatry, 165, 22–34. Post, F. (1996). Verbal creativity, depression and alcoholism: An investigation of one hundred American and British writers. The British Journal of Psychiatry, 168, 545–555. Raskin, A., Friedman, A. S., & DiMascio, A. (1982). Cognitive and performance deficits in depression. Psychopharmacology Bulletin, 18, 196–202. Roesch-Ely, D., Weiland, S., Scheffel, H., Schwaninger, M., Hundemer, H. P., Kolter, T., et al. (2006). Dopaminergic modulation of semantic priming in healthy volunteers. Biological Psychiatry, 60, 604–611. Roy, A., de Jong, J., & Linnoila, M. (1989). Cerebrospinal fluid monoamine metabolites and suicidal behavior in depressed patients: A 5-year follow-up study. Archives of General Psychiatry, 46, 609–612.
Schaffer, C. E., Davidson, R. J., & Saron, C. (1983). Frontal and parietal electroencephalogram asymmetry in depressed and nondepressed subjects. Biological Psychiatry, 18, 753–762. Servan-Schreiber, D., Printz, H., & Cohen, J. D. (1990). A network model of catecholamine effects: Gain, signal-to-noise ratio, and behavior. Science, 49, 892–895. Shimoda, K., & Robinson, R. G. (1999). The relationship between poststroke depression and lesion location in long-term follow-up. Biological Psychiatry, 45, 187–192. Starkstein, S. E., Robinson, R. G., Honig, M. A., Parikh, R. M., Joselyn, J., & Price, T. R. (1989). Mood changes after right-hemisphere lesions. British Journal of Psychiatry, 155, 79–85. Stuss, D. T., Alexander, M. P., Hamer, L., Palumbo, C., Dempster, R., Binns, M., et al. (1998). The effects of focal anterior and posterior brain lesions on verbal fluency. Journal of the International Neuropsychological Society, 4, 265–278. Vuga, M., Fox, N. A., Cohn, J. F., George, C. J., Levenstein, R. M., & Kovacs, M. (2006). Long-term stability of frontal electroencephalographic asymmetry in adults with a history of depression and controls. International Journal of Psychophysiology, 59, 107–115. Wiese, H., & Schweinberger, S. R. (2011). Accessing semantic person knowledge: Temporal dynamics of nonstrategic categorical and associative priming. Journal of Cognitive Neuroscience, 23, 447–459. Wood, A. G., Saling, M. M., Abbott, D. F., & Jackson, G. D. (2001). A neurocognitive account of frontal lobe involvement in orthographic lexical retrieval: An fMRI study. NeuroImage, 14, 162–169.