- Email: [email protected]
Olfactory sensitivity of rats with transection of the lateral olfactory tract
132
Brain Research, 616 (1993) 132-137 © 1993 Elsevier Science Publishers B.V. All rights reserved 0006-8993/93/$06.00
BRES 18991
Olfactory sensitivity of rats with transection of the lateral olfactory tract Burton M. Slotnick and Frances W. Schoonover Department of Psychology, The American University, Washington, DC 20016 (USA) and Institute of Psychiatry and Human BehaL,ior, University of Maryland Medical System, Baltimore, MD 21201 (USA)
(Accepted 2 February 1993)
Key words: Olfaction; Olfactory threshold; Lateral olfactory tract; Olfactory sensitivity; Anosmia
Rats with discrete transection of the lateral olfactory tract (LOT) were tested using operant conditioning and psychophysical methods for their amyl acetate intensity difference threshold and absolute detection threshold. Experimental rats performed as well as controls on the easiest problems of both threshold series but their intensity difference threshold was approximately 2.5 times as high as controls and their absolute detection threshold was approximately 2.25 orders of magnitude higher than controls. The deficit in sensitivity in both tests was related to the frontal level at which the tract was cut; rats with the most rostral transection had the greatest increase in threshold. The absolute detection threshold of rats with transection of the LOT was increased to that of normal human subjects tested with the same apparatus.
INTRODUCTION In a recent report, Slotnick and Schoonover 17, using an olfactometer and o p e r a n t discrimination tests, found that large lesions of the olfactory peduncle or olfactory cortex p r o d u c e d severe deficits in olfaction but discrete lesions of only one bulbofugal pathway had little or no effect in p e r f o r m a n c e of simple odor detection and o d o r discrimination tasks. Central to the present investigation is their finding that rats with complete transection of the lateral olfactory tract ( L O T ) had few deficits and, generally, their p e r f o r m a n c e was comparable to that of normal or sham lesioned controls. This o u t c o m e is surprising because the L O T is the major projection pathway of the olfactory bulb and the only direct input from the olfactory bulb to the piriform cortex, periamygdaloid cortex and entorhinal cortex. Further, many of the more deeply situated olfactory cortical neurons that provide impulses to the olfactory thalamocortical system derive their input directly or indirectly from the L O T 1°. Thus, one might expect that, at the very least, transection of the L O T should p r o d u c e deficits in odor sensitivity. T h e tests used in the Slotnick and Schoonover ~7 study were designed to detect residual olfactory function in animals with large lesions of olfac-
tory cortex and, hence, employed detection and discrimination problems that were relatively easy for normal rats. While the d e m a n d s of these tasks were not sufficient to detect the effects of transecting the LOT, other studies have reported that transection of the tract may p r o d u c e m a r k e d deficits in o d o r detection and discrimination. Thus L o n g and T a p p 7 found that rats with L O T lesions similar to those p r o d u c e d in the Slotnick and Schoonover ~7 study falied or m a d e many errors in an odor detection task and Cattarelli 2 reported that rats with similar lesions had a severe sensory deficit but were not anosmic. T h e reasons for these differing o u t c o m e s are unknown but one possibility is that they relate to differences in task difficulty. Studies which report marked deficits in odor detection from transection of the lateral olfactory tract may have employed a low odor concentration or one whose signal value was partly masked because of inadequate control of vapors in the test environment. Except for a few studies 1'3'11A5'16 most prior investigations of neuroanatomical correlates of olfactory function have not used quantitative olfactometric m e t h o d s or psychophysical tests and it is difficult at best to c o m p a r e or evaluate their outcomes. This problem is addressed in the present study by
Correspondence: B.M. Slotnick, Department of Psychology, The American University Washington, DC 20016, USA. Fax: (1) (202) 885-7617.
133
assessing the intensity difference and absolute detection threshold of rats in which the lateral olfactory tract was transected. MATERIALS AND METHODS
Subjects Eighteen adult male albino rats were used. T h e animals were maintained in a temperature- and humidity-controlled vivarium in individual plastic cages. In addition, 3 male and 4 female graduate and undergraduate students were tested on the absolute threshold task described below.
Apparatus The olfactometer and test c h a m b e r were identical to that described by Slotnick and Schoonover TM. Briefly, odors were generated in a positive-pressure, 5-stage air-dilution olfactometer, constructed entirely of glass and Teflon, that allowed odor concentration to be varied continuously from 100% of saturation to 10-7% of saturation. Each of two stimulus channels could be set to provide odorized or clean air. For absolute odor threshold tests, one channel provided the odor stimulus and the other channel provided clean air. For intensity difference threshold tests, one channel was set to provide a fixed concentration of the stimulus and the other provided a different concentration. Odor delivery was governed by electrically operated valves and all procedures were automated and controlled by an Apple lie computer connected to the olfactometer by a digital interface. All stimulus concentrations are reported as percent of vapor saturation at 20°C. The test chamber consisted of a modified glass funnel attached to a 13-cm-long Plexiglas tube. A perforated glass disc in the stem of the funnel diffused the incoming air stream. A stainless steel bar cemented through the left wall of the funnel served as a response key. Contacts between the bar and a stainless steel floor plate were detected by a touch-sensitive circuit 4. A photocell positioned across the neck of the chamber, 2 cm behind the air-diffusing screen, served to detect trial-initiating responses. W a t e r reinforcement from a solenoid-controlled reservoir was delivered to a glass cup at the base of the funnel. Air from the c h a m b e r was passed continuously to the outside of the building by a 70 cfm exhaust fan connected to the chamber by a flexible plastic hose. T h e test chamber was housed in a refrigerator that was thermostatically controlled to maintain the c h a m b e r temperature at 20°C.
Procedures Rats were restricted to 10 ml of water a day. This water deprivation schedule was initiated 14 days prior to training and was in effect throughout the study except for the first 3 days after surgery when ad lib. water was allowed. The deprivation schedule resulted in stabilization of body weight at 7 5 % - 8 0 % of weight prior to deprivation. Briefly, rats were initially trained to break the p h o t o b e a m at the neck of the chamber, sample the stimulus and then make contact with the response bar for a water reward. W h e n this training was completed, a go, no-go discrimination task was introduced. O n each trial either a positive (S + ) or negative ( S - ) stimulus was presented for 4 s. Only responses in the presence of the S + were reinforced. Additional details of the training and trial procedures are given by Slotnick and Schoonover 16.
Preoperative training Animals were given 800 trials (in 100-trial daily sessions) on a simple detection task in which a 0.5% iso amyl acetate vapor served as S + and clean air served as S - . All animals acquired this detection task in the first session and maintained stable performance of 9 0 - 1 0 0 % correct responding in the remaining trials. For the remaining tests, criterion performance was set at a m e a n of 75% or higher correct responding in the last 60 trials of the session.
Postoperative training Intensity difference threshold. In the first postoperative test, rats were tested on the 0.5% amyl acetate detection task as described above. They were then trained on an intensity difference discrimination task in which the S + stimulus (standard stimulus) was 0.1% amyl acetate and the S - stimulus (comparison stimulus) was 0.01% amyl acetate. Training was continued in daily 100-trial sessions (7 d a y s / w e e k ) until the rat reached criterion or for a m a x i m u m of 4 sessions (400 trials). If criterion performance was achieved, the concentration of the S stimulus was increased by 0.01% in the next session. W h e n the rat failed to reach criterion, training on the intensity difference task was terminated and the rat was trained on the absolute detection task described below. Absolute detection threshold. In the first 100-trial session the S + stimulus was 0.5% amyl acetate and the S - stimulus was clean air. After reaching criterion performance the concentration of the S + stimulus was decreased by 10 - ° 5 in the next session. This procedure was continued until the rat failed to reach criterion in 400 trials. The 7 h u m a n subjects were also tested on this task. Stimuli were: sampled from a 2.5-cm-diameter glass tube connected to output valves of the olfactometer. The subject sat in front of a large window exhaust fan and was provided with a trial-initiating key and a response key. The subject was instructed to initiate a trial whenever h e / s h e was ready and to press the response key if an odor was detected. A light signaled the onset of the stimulus. Subjects completed 60-100 trials in a session. Testing was terminated if response accuracy in the last 60 trials was below 75% correct. Except for these differences, the procedures and trial parameters were the same as those used for the rats. For both psychophysical tests the m e a n performance in the last 60 trials of each problem was plotted and a perpendicular line was drawn from the point at which the psychometric function crossed the 75% correct level to the value of the odor stimulus (absolute threshold) or comparison stimulus (intensity difference threshold) on the abscissa. T h e intensity difference threshold was computed using the formula ( S - C ) / S where S was the standard stimulus (S + ) and C was the value of the comparison stimulus determined by the graphic interpolation procedure. In addition, the absolute and intensity difference threshold of each subject was determined and these values were used for statistical analyses.
Surgery and histological control Rats were arbitrarily assigned to a s h a m lesioned control (n = 5), unilateral bulbectomy plus neocortical lesion control (n = 6) or unilateral bulbectomy plus lateral olfactory tract transection group (n = 7). Rats were anesthetized with 40 m g / k g of intraperitoneal sodium pentobarbital and clamped into a Kopf stereotaxic instrument. For sham-lesioned controls, the scalp was incised and the dorsal surface of the skull was exposed. For lesion controls, the left olfactory bulb was exposed and removed by aspiration u n d e r 20 × magnification. Special care was taken to insure that all nerve fascicles at the cribriform plate were severed and that the posterior medial part of the bulb, which extends under the frontal pole cortex, was removed. T h e anterolateral convexity of the right hemisphere was then exposed 9 and an #11 scalpel blade was used to make a 2 - 3 m m long, 3 - 4 m m deep incision in the cortex parallel to and just above the rhinal fissure. For experimental rats the surgical procedures were similar except that in the right hemisphere the lateral olfactory tract was visualized and transected using the #11 scalpel blade. In some cases a 1 - 2 m m segment of the tract and surrounding piriform cortex was also aspirated using a fine glass pipette. Scalp wounds were closed with metal clips and animals were maintained on a warm pad until they regained their righting reflex. T h e animals were given ad lib. water for the first 3 days after surgery and then returned to their water deprivation schedule. At the completion of behavioral testing rats were sacrificed u n d e r deep anesthesia by cardiac perfusion of saline followed by 10% formalin. The brains were stored in 10% formalin and 25% sucrose for 3 - 1 0 days and then sectioned at 50 p, on a freezing microtome. Every other section through the lesion was m o u n t e d on gelatincovered glass slides and stained with Cresyl violet. In some cases, alternate sections were stained with Sudan Black B for myelinated
134 fibers. The brain coordinates from the Slotnick and Hersch 12 stereotaxic atlas were used to define the frontal level of the lesions in experimental animals. The Student t-test was used for paired comparisons and a one-way A N O V A was used to evaluate differences among the three groups on the absolute threshold task.
RESULTS
Anatomical results Each of the lesioned control rats had the left olfactory bulb completely removed. The lesion extended into the anterior olfactory nucleus and, in two cases, into the rostral olfactory tubercle on the left. The rostral tip of the left frontal pole cortex was also damaged in most cases. On the right hemisphere there was a narrow 2 - 4 mm incision approximately 1-2 mm dorsal and parallel to the rhinal fissure in the lateral orbital cortex.
Rat
R22 -2.26* 0.80**
R21 -2.89 0.74
In experimental rats, the lesions in the left hemisphere were similar to those in the lesioned control group and in each rat the left olfactory bulb was completely removed. In the right hemisphere, the L O T was completely transected in each rat (Fig. 1). Among rats, lesions varied in size and with regard to the rostral level at which the tract was transected. In 3 cases (e.g. R22, Fig. 1) the lesions were very discrete and transected the tract with little damage to layer 2 of piriform cortex. In two other cases (e.g. R26, Fig. 1) the lesions were more extensive, destroyed piriform cortex at the level of the transection and extended into the lateral olfactory tubercle, the caudate nucleus and into cortex above the rhinal fissure. In the remaining 2 cases the lesions were confined to olfactory cortex but extended medially to invade the lateral 1 / 3 or 1 / 2 of the olfactory tubercle.
R26 -2.90 0.73
R20 -4.58 0.30
Frontal
15.0
13.6
13.2
12.3
~ 0
* Log absolute threshold ** Intensity difference threshold
Fig, 1. Diagrammatic representation of lesions in the right hemisphere of 4 experimental rats. In each case the right olfactory bulb was completely removed. Frontal levels refer to those given in the Slotnick and Hersh 12 atlas of the olfactory system. Rat R22 had the most anterior L O T transection of the group and R20 the most posterior transection. Note that both absolute threshold (*) and intensity difference threshold (**) are graded with respect to the rostral level at which the tract was transected.
135 The most rostral level at which the tract was completely transected (R22, Fig. 1) was at the posterior aspect of the anterior olfactory nucleus, at the most rostral level of the olfactory tubercle (frontal 14.2). The most posterior transection (R20, Fig. 1) was at the level of the genu of the corpus callosum (frontal level 12). Transection of the tract in the remaining rats (e.g. R21, R26, Fig. 1) was at the level of the rostral olfactory tubercle (frontal levels 13.5-12.3). Schematic diagrams of representative lesions shown in Fig. 1 are ordered with respect to the rostral level of the transection.
0.8"
0.7" 0.6"
0.5" 0.4" ¢0 0.3"
I
$ 0.2'
0.l'
0.0 LOT
Controls
Behavioral results There were no significant differences between sham lesioned and lesioned controls for either the intensity difference or absolute difference threshold measures and, for purposes of statistical analyses, these two subgroups were combined into a single control group. Intensity difference thresholds. The mean intensity dif-
Group -1'
-2'
V
-3'
100"
"~
90-
80-
-6 Controls
LOT
Human
Ss
cD L~
70"
0
Group
Controls
60. i
50
i
i
.1
.2
J
.3
.4 Delta
i
i
I
.5
.6
.7
.s
.9
UI
I00.
,~
90.
tt~ ~.~
80"
¢d I.i
70.
60.
$0
S U .
-6
.
-5'
.
.
-4'
Fig. 3. Top, mean amyl acetate intensity difference threshold (Delta I/I) for control rats and those with transection of the LOT. Bottom, mean amyl acetate absolute threshold for control rats, rats with transection of the LOT and human subjects. Solid circles in both graphs represent threshold scores of individual subjects. Log concentration is the log of percent of vapor saturation. See text for explanation of how threshold values were calculated.
(9
Controls
•
LOT
[]
Humans
i
-3'
-2'
-I
0
Log Concentration Fig. 2. Mean percent accuracy in the last 60 trials as a function of test condition. Top, amyl acetate intensity difference (Delta 1/I) threshold functions for control rats and those with transection of the LOT. Bottom, amyl acetate absolute intensity threshold functions for control rats, rats with transection of the LOT and human subjects. Vertical lines in both graphs represent 1 standard deviation. Log concentration is the log of percent of vapor saturation.
ference threshold of LOT rats (mean, 0.53) was significantly (P < 0.002) higher than that of controls (mean, 0.22). Experimental rats performed nearly as well as controls on the initial two intensity difference tasks (Fig. 2) but only the rat with the most posterior LOT transection reached criterion on the 0.1% vs. 0.06% problem while all controls reached criterion on the 0.1% vs. 0.07% problem (Fig. 3). Note that the mean intensity difference threshold for experimental rats based on the group psychophysical function (Fig. 2) is somewhat higher than the mean of the thresholds for individual rats. This is because the group psychophysical function is less influenced by the essentially normal performance of the one experimental rat with the most posterior lesion. Absolute detection threshold. The difference in threshold among the control (1.5 × 10-5), experimental (2.4 x 10 -3) and human groups (4.5 × 10 -3) was significant
136 ( P < 0.001; Fig. 2). The control rats had significantly lower thresholds than experimental rats or human subjects ( P < 0.02, each comparison) and, in fact, each of the human subjects and all but the experimental rat with the most posterior lesion had higher thresholds than those of control rats (Fig. 3). Thresholds of the human and L O T groups were not significantly different.
Lesion placement and odor sensitivity Experimental rats were rank ordered with respect to the anterior-posterior level of the L O T transection and Spearman rank order correlations between level of transection and threshold scores were computed. Rats with more rostral transections performed more poorly on the tests for intensity difference threshold (Rho = -0.72, P < 0 . 0 5 ) and absolute detection threshold (Rho = -0.89, P < 0.01). DISCUSSION The absolute detection threshold (1.5 x 10 -5) and intensity difference threshold (0.22) obtained for the 11 control rats in this study are essentially identical with thresholds (1.3 x 10 -5 and 0.31, respectively) obtained for 20 normal and unilateral bulbectomized rats in a prior study in this laboratory ~6. No differences were found between the performance of normal and unilateral bulbectomized control rats in either study. Contributing to this close replication in psychophysical data is the fact that the same olfactometer and test procedures were used in both studies. As described by Slotnick and Schoonover ~6, this absolute detection threshold is similar to or lower than those reported in other studies of olfactory threshold for amyl acetate in rats, rabbits, and dogs. Thus, our test procedures appear to be reliable and to yield a relatively sensitive measure of detection threshold. However, the intensity difference threshold obtained in these two studies was approximately an order of magnitude higher than that previously reported for the rat by Slotnick and Ptak 14. The difference among these studies is probably due, in part, to the much more extensive training given animals, the use of liquid dilutions of the odorant and the use of a staircase test procedure in the Slotnick and Ptak work. Thus, the intensity difference threshold obtained in this study may underestimate this capacity in control and experimental animals but the method is probably adequate to reflect the relative magnitude of the lesion effect. The results of this study confirm and extend the prior report of Slotnick and Schoonover 17 that transection of the L O T has little or no effect on the ability of
the rat to detect relatively high concentrations (0.5% and 0.05%) of amyl acetate or to discriminate a 90% difference in odor concentration. However, the psychophysical tests demonstrate that the lesion results in a decrease in odor sensitivity of approximately 2.25 orders of magnitude for detection and an approximate 2.5 fold decrease in ability to discriminate between concentrations of the same odor. Another index of the lesion-induced change in sensitivity is provided by comparing the performance of the macrosmatic rat to the microsmatic human. The absolute detection threshold of normal rats was approximately 2-3 orders of magnitude lower than that of humans and the effect of transection of the L O T was to eliminate this difference in sensitivity. Indeed, absolute detection thresholds for the 5 rats with the most rostral transection of the LOT were essentially identical to those of human subjects (Fig. 3). The intensity difference threshold for human subjects tested with amyl acetate was determined to be, on average, 30% by Slotnick and Ptak and this is appreciably better than the intensity difference threshold of rats with transection of the LOT. However, as discussed above, the methods used in the Slotnick and Ptak study differed in several important ways from those of this study and it is unclear if the absolute values for the intensity differences obtained in the two studies can be compared directly. In any case, it appears that olfactory sensitivity of rats with transection of the lateral olfactory tract is probably no better than that of the human. Thus, one major effect of divorcing the piriform and entorhinal cortices from direct projections of the olfactory bulb is a sharp decrease in olfactory sensitivity. Certainly, other olfactory functions, not assessed in this study, would also be compromised by L O T transection. As discussed by Slotnick and Schoonover ~7, the findings reported in other studies that L O T transection results in anosmia or severe hyposmia in the rat (e.g. refs. 2, 7, 20) probably stem from inadequate test procedures or the use of particularly difficult detection or discrimination tasks. Because experimental rats in this study had an intact anterior commissure (AC) they may have had olfactory input to the olfactory cortex in the control (bulbectomized) hemisphere. In unpublished studies we found that transection of the commissural part of the AC had little or no effect on simple odor detection and discrimination tasks in unilaterally bulbectomized rats with contralateral transection of the LOT. Further, Slotnick and Schoonover 17 found that the deficits in rats with bilateral olfactory cortical lesions were no greater than those with similar but unilateral lesions plus removal of the olfactory bulb in the contralateral
137 hemisphere. However, psychophysical tests were not used in that study and it is possible that olfactory information via the A C to the piriform cortex in the control h e m i s p h e r e m a y have contributed to olfactory sensitivity of experimental rats in the present study. A l t h o u g h little is k n o w n about the role of piriform cortex in olfactory behavior, o t h e r direct and indirect targets of the L O T have b e e n investigated and the olfactory function of these structures would be affected by transection of the lateral olfactory tract. Thus, the entorhinal cortex, which receives a direct projection from the L O T , has b e e n implicated in olfactory m e m ory 15'18'19. AlSO, the central segment of the mediodorsal thalamic nucleus, which receives olfactory projections from cells d e e p to the piriform cortex 1°, is known to play an important role in complex olfactory discrimination tasks 3'833. W h e t h e r these m o r e central olfactory sites contribute to olfactory sensitivity, as defined by the psychophysical measures used in the present experiment, is not known. Certainly it is not surprising that transection of a major bulbofugal p a t h w a y and the c o n s e q u e n t d e a f f e r e n t a t i o n of a large segment of olfactory cortex might p r o d u c e a decrease in olfactory sensitivity. However, the lesion certainly does not r e n d e r the animal anosmic and the present results, t o g e t h e r with those of our prior study 17' indicate that considerable olfactory function remains. To determine which bulbofugal projections may mediate this remaining function, Slotnick and S c h o o n o v e r a7 injected H R P into the olfactory bulb of rats with transection of the L O T and examined the distribution of a n t e r o g r a d e l y transported reaction product. T h e results indicated that the transection eliminated input f r o m the L O T to the piriform cortex but left intact bulbofugal terminations in the anterior olfactory nucleus and to virtually the entire olfactory tubercle. Thus, residual function is probably m e d i a t e d by the further projection of one or both of these structures. A n a t o m i c a l studies of H e i m e r and his associates 5'6 suggest that the olfactory tubercle may be a c o m p o n e n t of the ventral striatum and, as such, may have a quite different role than that of the piriform cortex in processing olfactory input. It would be of considerable interest to determine the contributions to olfactory function of these remaining projections. In summary, the present results indicate that transection of the lateral olfactory tract in the rat p r o d u c e s a significant d e c r e m e n t in absolute detection and intensity difference olfactory thresholds, reducing the sensitivity of this macrosmatic animal to approximately the level of the microsmatic human. However, and in a g r e e m e n t with Slotnick and S c h o o n o v e r ~7, the transection has little or no effect on the rat's ability to
p e r f o r m simple o d o r detection and discrimination tasks and these residual abilities are probably m e d i a t e d by the m o r e indirect projections from the anterior olfactory nucleus to piriform cortex a n d / o r the direct bulbofugal projections to the olfactory tubercle. This research was supported, in part, by MH Grant DC01266 to B.M.S.
Acknowledgements.
REFERENCES 1 Bennett, M.H., The role of the anterior limb of the anterior commissure in olfaction, Physiol. Behav., 3 (1968) 507-515. 2 Cattarelli, M., Transmission and integration of biologically meaningful olfactory information after bilateral transection of the lateral olfactory tract in the rat, Behav. Brain Res., 6 (1982) 313-337. 3 Eichenbaum, H., Shedlack, K.J. and Eckmann, K.W., Thalamocortical mechanisms in odor guided behavior. I. Effects of lesions of the mediodorsal thalamic nucleus and frontal cortex on olfactory discriminations in the rat, Brain Behav. Evol., 17 (1980) 255-275. 4 Field, B.F. and Slotnick, B.M., A multi-purpose photocell and touch circuit amplifier, Physiol. Behav., 40 (1987) 127-129. 5 Heimer, L., The olfactory cortex and the ventral striatum. In K. Livingston and O. Hornykiewicz (Eds.), Limbic Mechanisms. The Continuing Evolution of the Limbic System Concept, Plenum, New York, 1978, pp. 96-187. 6 Heimer, L., Zaborszky, L., Zahm, D.S. and Alheid G.F., The ventral striatopallidothalamic projection: I. The striatopallidal link originating in the striatal parts of the olfactory tubercle, J. Comp. Neurol., 255 (1987) 571-591. 7 Long, C.J. and Tapp, J.T., Significance of olfactory tracts in mediating response to odors in the rat, J. Comp. Physiol. Psychol., 72 (1970) 435-443. 8 Lu, X.M. and Slotnick, B.M., Acquisition of an olfactory learning-set in rats with lesions of the mediodorsal thalamic nucleus, Chem. Senses, 15 (1990) 713-724. 9 Powell, T.P.S., Cowan, W.M. and Raisman, G., The central olfactory connexions, J. Anat., 99 (1965) 791-813 10 Price, J.L and Slotnick, B.M., Dual olfactory representation in the rat thalamus: an anatomical and electrophysiological study, J. Comp. Neurol., 214 (1983) 63-77. 11 SIotnick, B.M., Olfactory Perception. In W. Stebbins and M. Berkley (Eds.), Comparative Perception. Vol. 1: Basic Mechanisms, Wiley, New York, 1990. 12 Slotnick, B.M. and Hersch, S., A Stereotaxic Atlas of the Rat Olfactory System, Brain Res. Bull., 5 (Suppl. 5), (1980) 1-55. 13 Slotnick, B.M. and Kaneko, N., Role of mediodorsal thalamic nucleus in olfactory discrimination learning, Science, 214 (1981) 91-92. 14 Slotnick, B.M. and Ptak, J.E., Olfactory intensity-difference thresholds in rats and humans, Physiol. Behav., 19 (1977) 795-802. 15 Slotnick, B.M. and Risser, J.M., Odor memory and odor learning in rats with lesions of the lateral olfactory tract and mediodorsal thalamic nucleus, Brain Res., 529 (1990) 23-29. 16 Slotnick, B.M. and Schoonover, F.W., Olfactory thresholds in normal and unilaterally bulbectomized rats, Chem. Senses, 9 (1984) 325-340. 17 Slotnick, B.M. and Schoonover, F.W., Olfactory pathways and the sense of smell, Neurosci. Biobehav. Rec., 16 (1992) 453-472. 18 Staubli, U., Fraser, D, Kessler, M. and Lynch, G., Studies on retrograde and anterograde amnesia of olfactory memory after denervation of the hippocampus by entorhinal cortex lesions, Behav. Neurol. Biol., 46 (1986) 432-444. 19 Staubli, U., Ivy, G. and Lynch, G., Hippocampal denervation causes rapid forgetting of olfactory information in rats, Proc. Natl. Acad. Sci. USA, 81 (1984) 5885-5887. 20 Thompson, R., Some subcortical regions critical for retention of an odor discrimination in albino rats, Physiol. Behav., 24 (1980) 915-921.
Brain Research, 616 (1993) 132-137 © 1993 Elsevier Science Publishers B.V. All rights reserved 0006-8993/93/$06.00
BRES 18991
Olfactory sensitivity of rats with transection of the lateral olfactory tract Burton M. Slotnick and Frances W. Schoonover Department of Psychology, The American University, Washington, DC 20016 (USA) and Institute of Psychiatry and Human BehaL,ior, University of Maryland Medical System, Baltimore, MD 21201 (USA)
(Accepted 2 February 1993)
Key words: Olfaction; Olfactory threshold; Lateral olfactory tract; Olfactory sensitivity; Anosmia
Rats with discrete transection of the lateral olfactory tract (LOT) were tested using operant conditioning and psychophysical methods for their amyl acetate intensity difference threshold and absolute detection threshold. Experimental rats performed as well as controls on the easiest problems of both threshold series but their intensity difference threshold was approximately 2.5 times as high as controls and their absolute detection threshold was approximately 2.25 orders of magnitude higher than controls. The deficit in sensitivity in both tests was related to the frontal level at which the tract was cut; rats with the most rostral transection had the greatest increase in threshold. The absolute detection threshold of rats with transection of the LOT was increased to that of normal human subjects tested with the same apparatus.
INTRODUCTION In a recent report, Slotnick and Schoonover 17, using an olfactometer and o p e r a n t discrimination tests, found that large lesions of the olfactory peduncle or olfactory cortex p r o d u c e d severe deficits in olfaction but discrete lesions of only one bulbofugal pathway had little or no effect in p e r f o r m a n c e of simple odor detection and o d o r discrimination tasks. Central to the present investigation is their finding that rats with complete transection of the lateral olfactory tract ( L O T ) had few deficits and, generally, their p e r f o r m a n c e was comparable to that of normal or sham lesioned controls. This o u t c o m e is surprising because the L O T is the major projection pathway of the olfactory bulb and the only direct input from the olfactory bulb to the piriform cortex, periamygdaloid cortex and entorhinal cortex. Further, many of the more deeply situated olfactory cortical neurons that provide impulses to the olfactory thalamocortical system derive their input directly or indirectly from the L O T 1°. Thus, one might expect that, at the very least, transection of the L O T should p r o d u c e deficits in odor sensitivity. T h e tests used in the Slotnick and Schoonover ~7 study were designed to detect residual olfactory function in animals with large lesions of olfac-
tory cortex and, hence, employed detection and discrimination problems that were relatively easy for normal rats. While the d e m a n d s of these tasks were not sufficient to detect the effects of transecting the LOT, other studies have reported that transection of the tract may p r o d u c e m a r k e d deficits in o d o r detection and discrimination. Thus L o n g and T a p p 7 found that rats with L O T lesions similar to those p r o d u c e d in the Slotnick and Schoonover ~7 study falied or m a d e many errors in an odor detection task and Cattarelli 2 reported that rats with similar lesions had a severe sensory deficit but were not anosmic. T h e reasons for these differing o u t c o m e s are unknown but one possibility is that they relate to differences in task difficulty. Studies which report marked deficits in odor detection from transection of the lateral olfactory tract may have employed a low odor concentration or one whose signal value was partly masked because of inadequate control of vapors in the test environment. Except for a few studies 1'3'11A5'16 most prior investigations of neuroanatomical correlates of olfactory function have not used quantitative olfactometric m e t h o d s or psychophysical tests and it is difficult at best to c o m p a r e or evaluate their outcomes. This problem is addressed in the present study by
Correspondence: B.M. Slotnick, Department of Psychology, The American University Washington, DC 20016, USA. Fax: (1) (202) 885-7617.
133
assessing the intensity difference and absolute detection threshold of rats in which the lateral olfactory tract was transected. MATERIALS AND METHODS
Subjects Eighteen adult male albino rats were used. T h e animals were maintained in a temperature- and humidity-controlled vivarium in individual plastic cages. In addition, 3 male and 4 female graduate and undergraduate students were tested on the absolute threshold task described below.
Apparatus The olfactometer and test c h a m b e r were identical to that described by Slotnick and Schoonover TM. Briefly, odors were generated in a positive-pressure, 5-stage air-dilution olfactometer, constructed entirely of glass and Teflon, that allowed odor concentration to be varied continuously from 100% of saturation to 10-7% of saturation. Each of two stimulus channels could be set to provide odorized or clean air. For absolute odor threshold tests, one channel provided the odor stimulus and the other channel provided clean air. For intensity difference threshold tests, one channel was set to provide a fixed concentration of the stimulus and the other provided a different concentration. Odor delivery was governed by electrically operated valves and all procedures were automated and controlled by an Apple lie computer connected to the olfactometer by a digital interface. All stimulus concentrations are reported as percent of vapor saturation at 20°C. The test chamber consisted of a modified glass funnel attached to a 13-cm-long Plexiglas tube. A perforated glass disc in the stem of the funnel diffused the incoming air stream. A stainless steel bar cemented through the left wall of the funnel served as a response key. Contacts between the bar and a stainless steel floor plate were detected by a touch-sensitive circuit 4. A photocell positioned across the neck of the chamber, 2 cm behind the air-diffusing screen, served to detect trial-initiating responses. W a t e r reinforcement from a solenoid-controlled reservoir was delivered to a glass cup at the base of the funnel. Air from the c h a m b e r was passed continuously to the outside of the building by a 70 cfm exhaust fan connected to the chamber by a flexible plastic hose. T h e test chamber was housed in a refrigerator that was thermostatically controlled to maintain the c h a m b e r temperature at 20°C.
Procedures Rats were restricted to 10 ml of water a day. This water deprivation schedule was initiated 14 days prior to training and was in effect throughout the study except for the first 3 days after surgery when ad lib. water was allowed. The deprivation schedule resulted in stabilization of body weight at 7 5 % - 8 0 % of weight prior to deprivation. Briefly, rats were initially trained to break the p h o t o b e a m at the neck of the chamber, sample the stimulus and then make contact with the response bar for a water reward. W h e n this training was completed, a go, no-go discrimination task was introduced. O n each trial either a positive (S + ) or negative ( S - ) stimulus was presented for 4 s. Only responses in the presence of the S + were reinforced. Additional details of the training and trial procedures are given by Slotnick and Schoonover 16.
Preoperative training Animals were given 800 trials (in 100-trial daily sessions) on a simple detection task in which a 0.5% iso amyl acetate vapor served as S + and clean air served as S - . All animals acquired this detection task in the first session and maintained stable performance of 9 0 - 1 0 0 % correct responding in the remaining trials. For the remaining tests, criterion performance was set at a m e a n of 75% or higher correct responding in the last 60 trials of the session.
Postoperative training Intensity difference threshold. In the first postoperative test, rats were tested on the 0.5% amyl acetate detection task as described above. They were then trained on an intensity difference discrimination task in which the S + stimulus (standard stimulus) was 0.1% amyl acetate and the S - stimulus (comparison stimulus) was 0.01% amyl acetate. Training was continued in daily 100-trial sessions (7 d a y s / w e e k ) until the rat reached criterion or for a m a x i m u m of 4 sessions (400 trials). If criterion performance was achieved, the concentration of the S stimulus was increased by 0.01% in the next session. W h e n the rat failed to reach criterion, training on the intensity difference task was terminated and the rat was trained on the absolute detection task described below. Absolute detection threshold. In the first 100-trial session the S + stimulus was 0.5% amyl acetate and the S - stimulus was clean air. After reaching criterion performance the concentration of the S + stimulus was decreased by 10 - ° 5 in the next session. This procedure was continued until the rat failed to reach criterion in 400 trials. The 7 h u m a n subjects were also tested on this task. Stimuli were: sampled from a 2.5-cm-diameter glass tube connected to output valves of the olfactometer. The subject sat in front of a large window exhaust fan and was provided with a trial-initiating key and a response key. The subject was instructed to initiate a trial whenever h e / s h e was ready and to press the response key if an odor was detected. A light signaled the onset of the stimulus. Subjects completed 60-100 trials in a session. Testing was terminated if response accuracy in the last 60 trials was below 75% correct. Except for these differences, the procedures and trial parameters were the same as those used for the rats. For both psychophysical tests the m e a n performance in the last 60 trials of each problem was plotted and a perpendicular line was drawn from the point at which the psychometric function crossed the 75% correct level to the value of the odor stimulus (absolute threshold) or comparison stimulus (intensity difference threshold) on the abscissa. T h e intensity difference threshold was computed using the formula ( S - C ) / S where S was the standard stimulus (S + ) and C was the value of the comparison stimulus determined by the graphic interpolation procedure. In addition, the absolute and intensity difference threshold of each subject was determined and these values were used for statistical analyses.
Surgery and histological control Rats were arbitrarily assigned to a s h a m lesioned control (n = 5), unilateral bulbectomy plus neocortical lesion control (n = 6) or unilateral bulbectomy plus lateral olfactory tract transection group (n = 7). Rats were anesthetized with 40 m g / k g of intraperitoneal sodium pentobarbital and clamped into a Kopf stereotaxic instrument. For sham-lesioned controls, the scalp was incised and the dorsal surface of the skull was exposed. For lesion controls, the left olfactory bulb was exposed and removed by aspiration u n d e r 20 × magnification. Special care was taken to insure that all nerve fascicles at the cribriform plate were severed and that the posterior medial part of the bulb, which extends under the frontal pole cortex, was removed. T h e anterolateral convexity of the right hemisphere was then exposed 9 and an #11 scalpel blade was used to make a 2 - 3 m m long, 3 - 4 m m deep incision in the cortex parallel to and just above the rhinal fissure. For experimental rats the surgical procedures were similar except that in the right hemisphere the lateral olfactory tract was visualized and transected using the #11 scalpel blade. In some cases a 1 - 2 m m segment of the tract and surrounding piriform cortex was also aspirated using a fine glass pipette. Scalp wounds were closed with metal clips and animals were maintained on a warm pad until they regained their righting reflex. T h e animals were given ad lib. water for the first 3 days after surgery and then returned to their water deprivation schedule. At the completion of behavioral testing rats were sacrificed u n d e r deep anesthesia by cardiac perfusion of saline followed by 10% formalin. The brains were stored in 10% formalin and 25% sucrose for 3 - 1 0 days and then sectioned at 50 p, on a freezing microtome. Every other section through the lesion was m o u n t e d on gelatincovered glass slides and stained with Cresyl violet. In some cases, alternate sections were stained with Sudan Black B for myelinated
134 fibers. The brain coordinates from the Slotnick and Hersch 12 stereotaxic atlas were used to define the frontal level of the lesions in experimental animals. The Student t-test was used for paired comparisons and a one-way A N O V A was used to evaluate differences among the three groups on the absolute threshold task.
RESULTS
Anatomical results Each of the lesioned control rats had the left olfactory bulb completely removed. The lesion extended into the anterior olfactory nucleus and, in two cases, into the rostral olfactory tubercle on the left. The rostral tip of the left frontal pole cortex was also damaged in most cases. On the right hemisphere there was a narrow 2 - 4 mm incision approximately 1-2 mm dorsal and parallel to the rhinal fissure in the lateral orbital cortex.
Rat
R22 -2.26* 0.80**
R21 -2.89 0.74
In experimental rats, the lesions in the left hemisphere were similar to those in the lesioned control group and in each rat the left olfactory bulb was completely removed. In the right hemisphere, the L O T was completely transected in each rat (Fig. 1). Among rats, lesions varied in size and with regard to the rostral level at which the tract was transected. In 3 cases (e.g. R22, Fig. 1) the lesions were very discrete and transected the tract with little damage to layer 2 of piriform cortex. In two other cases (e.g. R26, Fig. 1) the lesions were more extensive, destroyed piriform cortex at the level of the transection and extended into the lateral olfactory tubercle, the caudate nucleus and into cortex above the rhinal fissure. In the remaining 2 cases the lesions were confined to olfactory cortex but extended medially to invade the lateral 1 / 3 or 1 / 2 of the olfactory tubercle.
R26 -2.90 0.73
R20 -4.58 0.30
Frontal
15.0
13.6
13.2
12.3
~ 0
* Log absolute threshold ** Intensity difference threshold
Fig, 1. Diagrammatic representation of lesions in the right hemisphere of 4 experimental rats. In each case the right olfactory bulb was completely removed. Frontal levels refer to those given in the Slotnick and Hersh 12 atlas of the olfactory system. Rat R22 had the most anterior L O T transection of the group and R20 the most posterior transection. Note that both absolute threshold (*) and intensity difference threshold (**) are graded with respect to the rostral level at which the tract was transected.
135 The most rostral level at which the tract was completely transected (R22, Fig. 1) was at the posterior aspect of the anterior olfactory nucleus, at the most rostral level of the olfactory tubercle (frontal 14.2). The most posterior transection (R20, Fig. 1) was at the level of the genu of the corpus callosum (frontal level 12). Transection of the tract in the remaining rats (e.g. R21, R26, Fig. 1) was at the level of the rostral olfactory tubercle (frontal levels 13.5-12.3). Schematic diagrams of representative lesions shown in Fig. 1 are ordered with respect to the rostral level of the transection.
0.8"
0.7" 0.6"
0.5" 0.4" ¢0 0.3"
I
$ 0.2'
0.l'
0.0 LOT
Controls
Behavioral results There were no significant differences between sham lesioned and lesioned controls for either the intensity difference or absolute difference threshold measures and, for purposes of statistical analyses, these two subgroups were combined into a single control group. Intensity difference thresholds. The mean intensity dif-
Group -1'
-2'
V
-3'
100"
"~
90-
80-
-6 Controls
LOT
Human
Ss
cD L~
70"
0
Group
Controls
60. i
50
i
i
.1
.2
J
.3
.4 Delta
i
i
I
.5
.6
.7
.s
.9
UI
I00.
,~
90.
tt~ ~.~
80"
¢d I.i
70.
60.
$0
S U .
-6
.
-5'
.
.
-4'
Fig. 3. Top, mean amyl acetate intensity difference threshold (Delta I/I) for control rats and those with transection of the LOT. Bottom, mean amyl acetate absolute threshold for control rats, rats with transection of the LOT and human subjects. Solid circles in both graphs represent threshold scores of individual subjects. Log concentration is the log of percent of vapor saturation. See text for explanation of how threshold values were calculated.
(9
Controls
•
LOT
[]
Humans
i
-3'
-2'
-I
0
Log Concentration Fig. 2. Mean percent accuracy in the last 60 trials as a function of test condition. Top, amyl acetate intensity difference (Delta 1/I) threshold functions for control rats and those with transection of the LOT. Bottom, amyl acetate absolute intensity threshold functions for control rats, rats with transection of the LOT and human subjects. Vertical lines in both graphs represent 1 standard deviation. Log concentration is the log of percent of vapor saturation.
ference threshold of LOT rats (mean, 0.53) was significantly (P < 0.002) higher than that of controls (mean, 0.22). Experimental rats performed nearly as well as controls on the initial two intensity difference tasks (Fig. 2) but only the rat with the most posterior LOT transection reached criterion on the 0.1% vs. 0.06% problem while all controls reached criterion on the 0.1% vs. 0.07% problem (Fig. 3). Note that the mean intensity difference threshold for experimental rats based on the group psychophysical function (Fig. 2) is somewhat higher than the mean of the thresholds for individual rats. This is because the group psychophysical function is less influenced by the essentially normal performance of the one experimental rat with the most posterior lesion. Absolute detection threshold. The difference in threshold among the control (1.5 × 10-5), experimental (2.4 x 10 -3) and human groups (4.5 × 10 -3) was significant
136 ( P < 0.001; Fig. 2). The control rats had significantly lower thresholds than experimental rats or human subjects ( P < 0.02, each comparison) and, in fact, each of the human subjects and all but the experimental rat with the most posterior lesion had higher thresholds than those of control rats (Fig. 3). Thresholds of the human and L O T groups were not significantly different.
Lesion placement and odor sensitivity Experimental rats were rank ordered with respect to the anterior-posterior level of the L O T transection and Spearman rank order correlations between level of transection and threshold scores were computed. Rats with more rostral transections performed more poorly on the tests for intensity difference threshold (Rho = -0.72, P < 0 . 0 5 ) and absolute detection threshold (Rho = -0.89, P < 0.01). DISCUSSION The absolute detection threshold (1.5 x 10 -5) and intensity difference threshold (0.22) obtained for the 11 control rats in this study are essentially identical with thresholds (1.3 x 10 -5 and 0.31, respectively) obtained for 20 normal and unilateral bulbectomized rats in a prior study in this laboratory ~6. No differences were found between the performance of normal and unilateral bulbectomized control rats in either study. Contributing to this close replication in psychophysical data is the fact that the same olfactometer and test procedures were used in both studies. As described by Slotnick and Schoonover ~6, this absolute detection threshold is similar to or lower than those reported in other studies of olfactory threshold for amyl acetate in rats, rabbits, and dogs. Thus, our test procedures appear to be reliable and to yield a relatively sensitive measure of detection threshold. However, the intensity difference threshold obtained in these two studies was approximately an order of magnitude higher than that previously reported for the rat by Slotnick and Ptak 14. The difference among these studies is probably due, in part, to the much more extensive training given animals, the use of liquid dilutions of the odorant and the use of a staircase test procedure in the Slotnick and Ptak work. Thus, the intensity difference threshold obtained in this study may underestimate this capacity in control and experimental animals but the method is probably adequate to reflect the relative magnitude of the lesion effect. The results of this study confirm and extend the prior report of Slotnick and Schoonover 17 that transection of the L O T has little or no effect on the ability of
the rat to detect relatively high concentrations (0.5% and 0.05%) of amyl acetate or to discriminate a 90% difference in odor concentration. However, the psychophysical tests demonstrate that the lesion results in a decrease in odor sensitivity of approximately 2.25 orders of magnitude for detection and an approximate 2.5 fold decrease in ability to discriminate between concentrations of the same odor. Another index of the lesion-induced change in sensitivity is provided by comparing the performance of the macrosmatic rat to the microsmatic human. The absolute detection threshold of normal rats was approximately 2-3 orders of magnitude lower than that of humans and the effect of transection of the L O T was to eliminate this difference in sensitivity. Indeed, absolute detection thresholds for the 5 rats with the most rostral transection of the LOT were essentially identical to those of human subjects (Fig. 3). The intensity difference threshold for human subjects tested with amyl acetate was determined to be, on average, 30% by Slotnick and Ptak and this is appreciably better than the intensity difference threshold of rats with transection of the LOT. However, as discussed above, the methods used in the Slotnick and Ptak study differed in several important ways from those of this study and it is unclear if the absolute values for the intensity differences obtained in the two studies can be compared directly. In any case, it appears that olfactory sensitivity of rats with transection of the lateral olfactory tract is probably no better than that of the human. Thus, one major effect of divorcing the piriform and entorhinal cortices from direct projections of the olfactory bulb is a sharp decrease in olfactory sensitivity. Certainly, other olfactory functions, not assessed in this study, would also be compromised by L O T transection. As discussed by Slotnick and Schoonover ~7, the findings reported in other studies that L O T transection results in anosmia or severe hyposmia in the rat (e.g. refs. 2, 7, 20) probably stem from inadequate test procedures or the use of particularly difficult detection or discrimination tasks. Because experimental rats in this study had an intact anterior commissure (AC) they may have had olfactory input to the olfactory cortex in the control (bulbectomized) hemisphere. In unpublished studies we found that transection of the commissural part of the AC had little or no effect on simple odor detection and discrimination tasks in unilaterally bulbectomized rats with contralateral transection of the LOT. Further, Slotnick and Schoonover 17 found that the deficits in rats with bilateral olfactory cortical lesions were no greater than those with similar but unilateral lesions plus removal of the olfactory bulb in the contralateral
137 hemisphere. However, psychophysical tests were not used in that study and it is possible that olfactory information via the A C to the piriform cortex in the control h e m i s p h e r e m a y have contributed to olfactory sensitivity of experimental rats in the present study. A l t h o u g h little is k n o w n about the role of piriform cortex in olfactory behavior, o t h e r direct and indirect targets of the L O T have b e e n investigated and the olfactory function of these structures would be affected by transection of the lateral olfactory tract. Thus, the entorhinal cortex, which receives a direct projection from the L O T , has b e e n implicated in olfactory m e m ory 15'18'19. AlSO, the central segment of the mediodorsal thalamic nucleus, which receives olfactory projections from cells d e e p to the piriform cortex 1°, is known to play an important role in complex olfactory discrimination tasks 3'833. W h e t h e r these m o r e central olfactory sites contribute to olfactory sensitivity, as defined by the psychophysical measures used in the present experiment, is not known. Certainly it is not surprising that transection of a major bulbofugal p a t h w a y and the c o n s e q u e n t d e a f f e r e n t a t i o n of a large segment of olfactory cortex might p r o d u c e a decrease in olfactory sensitivity. However, the lesion certainly does not r e n d e r the animal anosmic and the present results, t o g e t h e r with those of our prior study 17' indicate that considerable olfactory function remains. To determine which bulbofugal projections may mediate this remaining function, Slotnick and S c h o o n o v e r a7 injected H R P into the olfactory bulb of rats with transection of the L O T and examined the distribution of a n t e r o g r a d e l y transported reaction product. T h e results indicated that the transection eliminated input f r o m the L O T to the piriform cortex but left intact bulbofugal terminations in the anterior olfactory nucleus and to virtually the entire olfactory tubercle. Thus, residual function is probably m e d i a t e d by the further projection of one or both of these structures. A n a t o m i c a l studies of H e i m e r and his associates 5'6 suggest that the olfactory tubercle may be a c o m p o n e n t of the ventral striatum and, as such, may have a quite different role than that of the piriform cortex in processing olfactory input. It would be of considerable interest to determine the contributions to olfactory function of these remaining projections. In summary, the present results indicate that transection of the lateral olfactory tract in the rat p r o d u c e s a significant d e c r e m e n t in absolute detection and intensity difference olfactory thresholds, reducing the sensitivity of this macrosmatic animal to approximately the level of the microsmatic human. However, and in a g r e e m e n t with Slotnick and S c h o o n o v e r ~7, the transection has little or no effect on the rat's ability to
p e r f o r m simple o d o r detection and discrimination tasks and these residual abilities are probably m e d i a t e d by the m o r e indirect projections from the anterior olfactory nucleus to piriform cortex a n d / o r the direct bulbofugal projections to the olfactory tubercle. This research was supported, in part, by MH Grant DC01266 to B.M.S.
Acknowledgements.
REFERENCES 1 Bennett, M.H., The role of the anterior limb of the anterior commissure in olfaction, Physiol. Behav., 3 (1968) 507-515. 2 Cattarelli, M., Transmission and integration of biologically meaningful olfactory information after bilateral transection of the lateral olfactory tract in the rat, Behav. Brain Res., 6 (1982) 313-337. 3 Eichenbaum, H., Shedlack, K.J. and Eckmann, K.W., Thalamocortical mechanisms in odor guided behavior. I. Effects of lesions of the mediodorsal thalamic nucleus and frontal cortex on olfactory discriminations in the rat, Brain Behav. Evol., 17 (1980) 255-275. 4 Field, B.F. and Slotnick, B.M., A multi-purpose photocell and touch circuit amplifier, Physiol. Behav., 40 (1987) 127-129. 5 Heimer, L., The olfactory cortex and the ventral striatum. In K. Livingston and O. Hornykiewicz (Eds.), Limbic Mechanisms. The Continuing Evolution of the Limbic System Concept, Plenum, New York, 1978, pp. 96-187. 6 Heimer, L., Zaborszky, L., Zahm, D.S. and Alheid G.F., The ventral striatopallidothalamic projection: I. The striatopallidal link originating in the striatal parts of the olfactory tubercle, J. Comp. Neurol., 255 (1987) 571-591. 7 Long, C.J. and Tapp, J.T., Significance of olfactory tracts in mediating response to odors in the rat, J. Comp. Physiol. Psychol., 72 (1970) 435-443. 8 Lu, X.M. and Slotnick, B.M., Acquisition of an olfactory learning-set in rats with lesions of the mediodorsal thalamic nucleus, Chem. Senses, 15 (1990) 713-724. 9 Powell, T.P.S., Cowan, W.M. and Raisman, G., The central olfactory connexions, J. Anat., 99 (1965) 791-813 10 Price, J.L and Slotnick, B.M., Dual olfactory representation in the rat thalamus: an anatomical and electrophysiological study, J. Comp. Neurol., 214 (1983) 63-77. 11 SIotnick, B.M., Olfactory Perception. In W. Stebbins and M. Berkley (Eds.), Comparative Perception. Vol. 1: Basic Mechanisms, Wiley, New York, 1990. 12 Slotnick, B.M. and Hersch, S., A Stereotaxic Atlas of the Rat Olfactory System, Brain Res. Bull., 5 (Suppl. 5), (1980) 1-55. 13 Slotnick, B.M. and Kaneko, N., Role of mediodorsal thalamic nucleus in olfactory discrimination learning, Science, 214 (1981) 91-92. 14 Slotnick, B.M. and Ptak, J.E., Olfactory intensity-difference thresholds in rats and humans, Physiol. Behav., 19 (1977) 795-802. 15 Slotnick, B.M. and Risser, J.M., Odor memory and odor learning in rats with lesions of the lateral olfactory tract and mediodorsal thalamic nucleus, Brain Res., 529 (1990) 23-29. 16 Slotnick, B.M. and Schoonover, F.W., Olfactory thresholds in normal and unilaterally bulbectomized rats, Chem. Senses, 9 (1984) 325-340. 17 Slotnick, B.M. and Schoonover, F.W., Olfactory pathways and the sense of smell, Neurosci. Biobehav. Rec., 16 (1992) 453-472. 18 Staubli, U., Fraser, D, Kessler, M. and Lynch, G., Studies on retrograde and anterograde amnesia of olfactory memory after denervation of the hippocampus by entorhinal cortex lesions, Behav. Neurol. Biol., 46 (1986) 432-444. 19 Staubli, U., Ivy, G. and Lynch, G., Hippocampal denervation causes rapid forgetting of olfactory information in rats, Proc. Natl. Acad. Sci. USA, 81 (1984) 5885-5887. 20 Thompson, R., Some subcortical regions critical for retention of an odor discrimination in albino rats, Physiol. Behav., 24 (1980) 915-921.