Effects of Lithium Chloride Induced Reward Devaluation on Two Types of Spatial Behavior of the Taxon System
20
0
0
全文
(2) 244. Jian- You Lin and Ruey-Ming Liao. Recent research in spatial behavior seems to indicate that spatial behavior is not a unitary sys tem; instead, it is a multiple system controlled by separate neural substrata. On the basis of spatial target setting, O'Keefe and Nadel (1978) divided spatial behavior into two main categories: (a) the locale system which is based on place hypothesis that postulates the construction of an internal rep resentation of the external spatial environment, and (b) the taxon system which is based on bodily orientation. The taxon system can be further di vided into two sub-systems: one uses spatial cues to guide behavior (the guidance strategy), and the other uses bodily orientation to achieve the same goal (the orientation strategy). When spatial cues are used, a target object in the environment serves normally as an anchor point to allow the organ ism to organize proper avoidance or approach be havior. When bodily orientation is involved in spatial behavior, the organism uses an egocentric spatial framework that allows the rotation of its own body to move in the proper detection. The e gocentric spatial framework may be based on the organism's eyes, head or body as the spatial axis. To reach its goal, the organism seems to learn to rotate its body a certain amount (e.g. 90°) along the axis. Thus this orientation strategy is often called egocentric spatial behavior. Research on the neural basis of spatial be havior has discovered that the two categories of spatial behavior proposed by O'Keefe and Nadel (1978) were indeed mediated by two separate neural mechanisms. The locale system depends on the hippocampal formation for its proper func tion. Any damage on, or interference with, this formation is known to seriously disrupt spatial behavior based on the locale system (Jarrard, 1983; Jarrard, Okaichi, Steward, & Goldschmidt, 1984; Kesner, Bolland, & Vakis, 1993; Knowlton, Shapiro & Olton, 1989; Morris, Garrud, Rawlins, & O'Keefe, 1982; Okaichi & Oshima, 1990; Olton & Samuelson, 1976). On the other hand, the neu ral mechanism of spatial behavior based on the taxon system is much less clearly understood than that of the locale system. Although it was pro posed that the taxon system is mediated by the. striatum or the caudate nucles, experimental demonstrations of this link are controversial. Pisa and Cyr (1990) found that caudate lesions affect ed the turn-discrimination task, but had no visible effect on the brightness-discrimination task. Mitchell and Hall (1988) reported that rats with caudate lesions performed poorly in a Y maze. Nevertheless, if an additional clearly marked spa tial object cue was provided, the rats could still learn to spatially discri minate the two arms of a Y maze. These studies suggest that lesions on cau date nucles result in the selective disruption of spatial behavior based on bodily orientation. However recent studies (Lyford and Jarrard, 1991; Packard, Hirsh, and White, 1989; Packard and McGaugh, 1994) found that caudate lesions can also affect spatial behavior based on spatial cues. These conflicting reports may indicate that the two strategies applied in the taxon system of spatial behavior depend on the caudate nucles to different extent. Given the lack of a specific learning compo nent, O'Keefe and Nadel (1978) pointed out that bodily orientation strategy may be relevant to re sponse learning while cue guidance strategy may be acquired through the association between a target goal and environmental anchor objects. However, Sage and Knowlton (2000) reported that performance in the win-stay task in maze learning (originally thought to depend on guid ance strategy) seems to depend on specific re sponse learning as well. Their results did not a gree with O'Keefe and Nadel's (1978) contention that stimulus-stimulus association is the only type of learning process involved in the guidance strategy. Therefore, both the neurological data and the learning-acquisition data speak against a single hypothetical process for the taxon system of spatial behavior. The present study aims to further clarify the underlying behavioral component of the taxon system by investigating the learning process in volved in this kind of spatial behavior. Specifi cally, behavioral components acquired through two different strategies will be compared to see if potentially different learning associations are in-. a. a. 11. S. d J€ il Ii u. ti l'.
(3) Effects of Lithium Chloride Induced Reward Devaluation. 1tal 'isa ~ct. ble ok. 'ith ze. pa till. lY au of ill.. 'd, lfd ms :ial hat of to. )0. lat reay. la. .s. ed ze d-. 'e. a. }n. ly ce ta. a m. Ie. III. I. I-. ;h if. 1-. volved. Learning theories generally divide learning into three types of associations: S-R, S-S* and R-S*, while S, Rand S* symbolizes environmen tal stimulus, response, and reward respectively. S-R associations contend that responses are pas sively induced by environmental stimulus, and furthermore, the strength of the induced response is a function of the strength of the physiological drive of the organism (Hull, 1943). S-S* and R-S* associations, on the other hand, require the con cept of expectancy to explain this type of learn ing. S-S* represents the capacity to anticipate S* given S, R-S* represents the anticipation of S* following R. The strength of the response in both cases depends on the incentive values of the S* (Bolles, 1972). Accordingly, we may determine the nature of an associative learning by answers to two questions: 1) does the strength of the in centive value influence response strength? And 2) do environmental stimuli determine the onset of a response? The first question deals with the relationship between incentive value and response strength. Given that the difference between S-R learning and S-S*, R-S* learning resides in anticipated re ward, one may infer that lowering the incentive value of a reward should affect responses learned through S-S* and R-S* associations, but not S-R responses. Lithium chloride devaluation is a commonly used method to lower the reward value ofa reward. When lithium chloride injections are paired with an incentive, taste aversion is associ ated with the incentive through conditioning, thereby lowering the incentive value of reward (Garcia, Kime1dorf & Koelling, 1955). Colwill and Rescorla (1985) used this procedure to exam ine the nature of learning of bar-pressing re sponses. They found that bar-pressing response decreased noticeably under lithium chloride in jection demonstrating the role of S* in the learn ing of bar-pressing. Previous studies using the lithium chloride injection method often exam ined the role of positive reward in operant condi tioning (Chen and Amsel, 1980; Dickinson, 1997); however, no study used this method to ex. 245. amine the role of incentives in constructing spa tial representation (e.g. in a radial maze). Howev er, Sage and Knowlton (2000) did use this method to examine the nature of learning involved in the win-shift and win-stay tasks with radial arms. Normally a place orientation strategy is required ofthe organism in the win-shift task, while in the win-stay task, spatial cue guidance strategy is re quired. Three steps were involved in their devalu ation method: 1) response learning, 2) lithium chloride injection, and 3) testing the effects of lowering incentive value on the learned response. Results from their study showed that lithium chloride injection increased the response latency in the win-shift task, but not in the win-stay task. Response error rate was not affected in either task. Thus it seems that S* is a critical factor in learning the win-shift task which must involve in centive related associative learning. On the other hand, in the win-stay task, S* did not seem to mat ter. Sage and Knowlton (2000) demonstrated the viability of using lithium chloride injection to study spatial learning. However, even though their results indicated the critical role of S * in the won-shift task, they failed to reveal whether S-S* learning or R-S* learning was responsible for the results observed in the win-shift task. The second question mentioned above con cerns whether environmental stimulus is essen tial for the execution of spatial behavior. If changing environmental stimulus altered spatial behavior, then S-R or S-S* learning is likely in volved in that behavior; otherwise, R-S* learning may be responsible for that behavior. In addition, given that spatial behavior is affected by environ mental stimulus, one needs to further examine the role of S-R or S-S* association in that spatial be havior. To determine what type of associative learn ing is involved in the spatial behavior of the taxon system, the present investigation manipulated both the environmental stimulus and the incen tive value of reward. Two types of radial arm maze were used in a cue-learning task and an ego centric task. In the former task, a single cue (sandpaper) was paired with reward in the baited.
(4) .Tian-You Lin and Ruey-Ming Liao. 246. arm. This reward cue was neither placed on a fixed arm of the maze, nor did it have an invariant spatial relation to the subject. Therefore the sub ject can only rely on the cue of entering the baited arm to guide its spatial behavior (Lyford & Jar rard, 1991; Okaichi & Oshima, 1990). In the ego centric task, on the other hand, the only way to obtain the reward is the subject's bodily orienta tion. Thus the subject must use bodily orientation to guide this type of spatial behavior (Cook & Kesner, 1988; Pisa & Cyr, 1990). Our predictions were that: 1) if lithium chlo ride injection does affect subjects' spatial behav ior, then that behavior must be learned through S-S* or R-S*, otherwise it is S-R learning. More over, 2) if altered environmental stimulus affects subjects' spatial behavior, then the task must in volve S-S* learning; otherwise, it must be R-S* learning.. Method Subjects. Eight-week old male Wi star rats were used as subjects. After they were purchase from Center for Experimental Animals of the Medical School of the National Taiwan Universi ty, each rat was housed individually in a stainless steel cage. The ambient temperature of the animal room was maintained at 23°e ± 1°e. Light/dark cycle was set at 12112 hours (7:30 to 19:30 as the light portion). Behavior test was conducted be tween 13:00 to 17:00 hours. Following free access to food and water for two weeks, a food depriva tion regimen was conducted throughout the ex periment until the weight of the rats came down to about 85% of their pre-deprivation level. Ex periments commenced at this body weight, which was maintained (with 14 grams of daily feed) un til the end of this study. Animal care and experi mentation of this study followed the guide speci fied in the "Ethical Standard of Psychological Professionals" published by the Chinese Psycho logical Association (1996). Apparatus. Both the cue-learning task and the egocentric task were conducted in a eight-arm radial maze, made of acrylate, placed on a table. (about 80cm above ground level) in the center of the test room. In the center of the maze, stood an octagonal platform (30cm in diameter) from which radiated eight runways (45° between adja cent runways), each of which measured 61cm long, 7.5 cm wide, and 15 cm in height. A small feeding cup was attached outside the end of each runway. Each unit of bait provided (in the feed cup) consisted of a grain of chocolate at 350-400 mg. Drugs. Lithium chloride (Sigma, St. Louis, Mo. USA) was dissolved in 0.9% saline solution to a concentration of 0.15 M. Each injection of lithium chloride was given intraperitoneally at 10 ml per one kg of body weight.. Procedures for maze training The cue learning task. Before training be gan, there was a two-day period of adaptation which consisted of placing the subjects in the maze without rewards for five minutes each day, to allow the subjects to explore the maze freely. After this period of adaptation, the experiment proper began. Reward was placed in the feed box of two randomly chosen runways that was marked by black coarse sandpaper (Cwl00 measuring at 28cm x 70.5cm) affixed at the entrance of the se lected runway. For each trial (only one trial was given each day) only two runways thus marked contain re. ward, and the baited arms were chosen according to a pseudorandom schedule that precluded two adjacent arms to be chosen at the same time. At the beginning of each trial, each subject was placed on the platform at the center of the radial maze with its head pointed in a fixed direction. There was a transparent acrylate fence that keeps the rat on the platform for five seconds. This fence was then removed to allow the rat to search for food. A trial terminated either after the rat consumed the two baits or after five minutes of fruitless search by the rat. The experimenter recorded the sequential order in which the rat en tered the arms, and the time to complete the trial. A criterion for learing was set for three consecu. j t. c t c r. n. tl. tl Ii. '"'. T. cl. m gi tr tr in. pE til. la he th jel re mi.
(5) Effects of Lithium Chloride Induced Reward Devaluation. ~r. of dan rom dja1cm naIl ~ach. °eed 400. uis, tion 1 of t 10. be :lOn the. lay, ~ly.. lent box ked ~ at se-. ach re mg :wo At ,vas hal on. eps his rch rat of lter en ial. cu. tive days of three or less arm entries to complete the task. The subjects (n = 31) reached the learning criterion after about 27 days of training. They were then subjected to the response probe test. This test consisted of two trials on separate days. To avoid the interference of consummatory be havior, no bait was given during the test. This meant that the test was conducted as an extinction trial. One of the two trials was a simple extinction trial which served as a base-line for comparison to the other trial which was a no cue trial (i.e. sandpaper was removed). For the no-cue trial, the subject was placed in the central platform and al lowed to explore the maze until it made three en tries; at which point, the test was terminated. In addition to the number of entries, the experi menter also recorded the amount of time the sub ject spent on the platform before choosing to en teran arm.. After the no-cue probe test, the subjects re ceived three additional days of retraining (with bait). When behavior reached the level of previ ous learning criterion, a single extinction trial (no reward, but with sandpaper) would follow on the next day. Similarly, the test was terminated after the subject chose to enter three arms. A separate group of rats which had reached the learning criterion of the cued task was given lithium chloride inj ections. This group of rats was divided into three subgroups (n = 8 each). The first subgroup (the paired group) had lithium chloride injection paired with reward. Fifteen minutes after eating 3g of chocolate the rats were given lithium chloride injection through the in traperitonal route. A second subgroup (the con trol group) was given injection of saline solution instead oflithium chloride. A third group (the un paired group) was given lithium chloride injec tion, but the injection was not paired with choco late consumption. The injection was given 20 hours after the intake of chocolate for subjects in the unpaired group. Usually after four or five in jections, the paired group would show apparent reluctance to eat chocolate. Injection would ter minate when this reaction became apparent.. 247. Following the devaluation procedure, sub jects were allowed two days ofrest (no injection, no training) before underwent the next stage of the experiment which consisted of four test trials given on four separate days. On the first and the fourth day, extinction trials were given; on the second and the third day retraining trials were given. The extinction trial was identical to the re training trials except that there was no bait. The same type of information as in training trials was recorded, except that time to task completion was divided into four temporal zones: 1) the time be tween removal of the plastic fence to entry of the first arm, 2) the time from entering the first arm to arriving at the feed cup, 3) the time from arrival to the first feed cup to entry of the second arm, and 4) the time from entering the second arm to ar rival at the feed cup. The egocentric task. After two days of adap tation (as described in the previous task), rats were tested on their preferred directions of turns in the maze on the third day. After this test, four of the eight arms of the maze were closed off by placing acrylic dates at the arm entrances, result ing in a cross-shaped ("+") maze that was used in subsequent tasks. Direction-preference test con sisted of three trials. At the beginning of each tri al, the rat was first placed at the end of an arm facing the central platform. When the rat went from this arm to another arm, it was immediately taken from the maze and returned to its home cage. Preferred direction of turn was then record ed. After the direction preference test, the rat was trained on the egocentric task. In this task, a ran domly chosen arm was used as the rat's starting point. Ninety degrees from the starting arm, and against the rat's preferred direction, another arm was baited with a chocolate reward at the end of the arm. The rat was placed at the end of the first chosen arm to start this test. The time from this moment on until it reached the end of the baited arm was recorded. Then the rat was removed from the maze. If the rat chose to enter the baited arm, the trial was classified as a correct trial; other wise it was an error. In either case, the recorded time did not include the time of consuming the.
(6) Jian-You Lin and Ruey-Ming Liao. 248. chocolate. There were five training everyday for as long as it took to reach three consecutive days of 80% (or more) of correct response rate. Then the training was terminated. After 11 days of training and meeting the baseline criterion, the subjects (n = 36) were giv en a probe response test. This test was identical to the training trial, except that the four arms previ ously blocked were now unblocked and the arms previously used in the training trials were now blocked. This procedure resulted in a new cross-shaped ("+") maze. This probe test was con ducted on a single day with five trials. The same types of data as in the training test were taken. Another batch of rats, trained with egocen tric task, was divided into three groups. They re ceived lithium chloride injection treatment iden tical to the procedures described before (with n = 9 for paired, n = 6 for unpaired, and n = 8 for con tro). After the reward devaluation procedure, a four-day test took place with five trials per day. In the first and the fourth day, extinction trials were given. On the second and the third day, retraining trials were given. Same data were taken as before. Data were analyzed with canned statistical packages "statistica" (Borland International Inc).. Results The cue-learning task Fig. 1 presents data for time spent on the platform in the cue-learning task when cues were removed as compared to intact cues. Time spent on the platform was significantly higher (F (1,30) = 16.87, p < .001) when cues were removed. This significant difference indicated that the subjects did learn to use sandpaper as the spatial cue to guide their spatial behavior. When this spatial cue was removed, subjects had difficulty to locate the chocolate reward and hesitated in their choice, thereby increased the time spent on the platform. Fig. 2 presents data on the effects ofthe lithi um chloride devaluation on the cue-task. Two-way (mixed) ANOVA performed on the. ~. 25. ~. '". '-'. ..e 20. -. ....0 = ~. -..=. ***. 15. -; 10 ~. U. 5. = 0. ~. .-e f. 0. NO BAIT NO CUE. NO BAIT. Figure 1. The time spent on the central platform measured for the probe test of the cue task, in which the subjects (n = 31) were tested under con ditions of the removal of both cue and bait (NO CUE + NO BAIT) as well as the removal of bait on ly (NO BAIT). Each bar is expressed as the mean 1 ± s.e.m. *** indicates a significant difference between the two probe condition. mean number of entries shown in the top panel of Fig. 2, revealed that there was no significant dif ference on any of these tests (ps > .05). For the time to complete the task shown in the bottom panel ofFi. 2, ANOVA showed significant differ ence. Main effects on Session (F (4,84) = 16.05, P < .001), Group (F (2,21) = 3.88, P < .05), and their interaction (F (8,84) = 3.03, p < .01) were all sig nificant. Simple main effect analyses on each ses sion indicated that only the first retraining (RETl) showed significant difference among the three groups (F (2, 21) = 5.88, P < .01). No signifi cant difference between these groups was found under any other condition (ps > .05). Post hoc comparisons on RETI with Scheffe's test revealed that the paired group had significantly higher val. ue than the control group (F (2, 21) = 10.52, P <. .05). When the paired group was compared with the unpaired injection group, the difference ap proached the .05 level of significance (F (2, 21) = 6.70,p = .055). Figure 3 presents detailed tempo ral information on these three groups under first retraining (RETl). Since the effect of lithium chloride injection made a difference between the three subgroups, only time spent in four different. F. o tr T. ril UI. 10. PE. th tic se pn 1~ litl". grc ZOI. wa difi terr test. Sigl. con ana..
(7) Effects of Lithium Chloride Induced Reward Devaluation. • Paired. rL1. ~Unpajred. ~ 10 Z. DContro1. 8. <j. Q,j. 30. f;I;o1. ~. 6. 20. f. ~ 4. f;I;l. lO. ==. ~ 2 ~. Z. *. 10 ~. ~. f;I;l. o~. • Paired ii1§]Unpaired DControl. 50. 12 ~ Eo<. 249. (). 0. 1. 3. 2. 4. ZONE ~ 100. Figure 3. The time spent in four different zones in the maze of the cue task. Further analy ses for the temporal data on the first retraining session after lithium chloride devaluation treat ment. (as the RET1 shown in the bottom panel of Figure 2). * indicates a significant difference be tween lithium chloride paired group and the saline control group.. ~. ~ 80. ...<. ). = 60 ~. ..l. ,. 3. ~ 40. 0. t.l. ...o ~. ~. f. e. n. '. r. .r. ,c_. :. g e. 1. d c. d l-. h. ,. ,. ;t. n Ie. 1t. ........ 20 0. BSL. EXTl. RETl. RET2. EXT2. Figure 2. Effects of lithium chloride devaluation on the cue task as measured by the number of en tries (top) and the time to complete task (bottom). Three groups (n 8 each) were: the lithium chlo ride pairing treatment (Paired), the lithium chloride un-pairing treatment (Unpaired), and the physio logical saline pairing treatment (Control). The ex perimental sessions include the baseline before the lithium chloride devaluation (BSL), two extinc tion sessions (EXT1 and EXT2), and two retraining sessions (RET1 and RET2). See text for detailed procedures. Each bar is expressed as the mean ± 1 s.e.m. * indicate a significant difference between lithium chloride paired group and the saline control group.. =. zones were further statistically analyzed. A one way ANOYA showed that there was a significant difference between the groups only in the third temporal zone, (F (2,21) = 5.24, p < .05). Scheffe's test further revealed that the paired group spent significantly longer time in this zone than the control group (F (2,21) = 9.99, p < .05). From the analyses described above, it seems that lithium. chloride devaluation did not affect the rate of cor rect response, but it did affect the time to com plete the task. This significant effect can be at tributed to the fact that lithium chloride devalua tion had significantly increased the time spent in the third zone by the paired group (including the time to eat the chocolate) compared to the control groups. Thus, the effect of lithium chloride deval uation seems confined to the consummatory be havior of the subjects. This treatment did not seem to affect spatial behavior as far as correct choice to enter baited arms is concerned.. The egocentric task Fig. 4 presents data of the probe test on the e gocentric task. The upper panel shows the correct response rate of the baseline (BSL) and the shift condition. There was no significant difference between these two conditions (p> .05). The lower panel shows the mean time to complete the task for baseline and shift condition. Response time between these conditions did show significant difference (F (1, 35) = 4.61, p < .05). These results indicated that the subject learned to reach its goal by bodily orientation, while shifting the starting.
(8) Jian- You Lin and Ruey-Ming Liao. 250. t)[;oil 100 0:::. 0:::. o 80. U. zE--[;oil 60 ~ 40 [;;oJ. Q.. Z. 20. [;;oJ. ~ 0 8 ,-.,. -; ...... ...."". 6. *. -y. ~. '". '-'. [;;oJ. 4. :t .... E- Z. 2 < [;oil ~. o SHIFf. .BSL. Figure 4. Mean percent of correct response (top) and mean time to complete a trial (bottom) mea sured for the probe test of the egocentric task, in which the subjects (n 36) were tested under con ditions of the baseline (BSL) and the gO-degree shift of starting point of the maze (SHIFT). Each bar is expressed as the mean ±1 s.e.m. * indicates a significant difference between two the probe condtion.. =. t'oint did not affect the rate of correct response. On the other hand, significantly longer response time under shift condition indicated that the ex perimental manipulation (i.e. shift condition) did increase the time for the rats to find their goals, thereby demonstrating the effectiveness of this manipulation. Fig. 5 presents data of the effects of incentive devaluation on the egocentric task. There were two retraining sessions (RETl and RET2), two. extinction sessions (EXT! and EXT2) and a base line session (BSL). The top panel presents data of the three treatment groups (paired, unpaired and control) across all five sessions. A two-way ANOVA revealed that both main effects were significant (F (2, 20) = 5.54, p < .05, for group ef fect; F (4, 80) = 2.66, p < .05 for session effect), but their interaction was not significant (p > .05). Follow-up analyses for simple main effects found that neither the first extinct nor the first retrain ing showed any significant difference between the three groups (ps > .05). However, in the sec ond retraining session, a significant effect (F (2, 20) = 4.54, p <.05) was found. In the second ex tinction session, a marginal significant effect (F (2, 20) = 3.36, p <.055) was detected. Post hoc comparison based on Scheffe's test found that the paired group had significant lower correct re sponse rate than the control group in the second retraining (F (2,20) = 9.0 I, P < .05) . Shown in the lower panel of Figure 4 was the mean time to complete the task. A two-way ANOVA found significant main effect for Session and for Session x Group interaction (F (4, 80) = 4.64, P < .0 I; F (8, 80) = 4.32, P < 0.001, respectively). The Group main effect failed to reach significance (p > .05). Further analyses showed no significant difference between the groups under the first retraining and the first extinction conditions (ps > .05), but under both the second retraining and the second extinction, the groups were significantly different (F (2, 20) = 4.63, P < .05 and F (2, 20) = 3.57, P < .05 respec tively). Post hoc comparisons for the second re training condition found that the paired group had significantly longer task completion time (F (2, 20) = 7.13, P < .05), the same was true for the sec ond extinction session (F (2, 20) = 10.84, P < .05). These data indicated that lithium chloride treatment decreased the rats' correct response as well as increased the time to complete the ego centric task. The lack of significant effects for the first extinction and the first retraining session probably meant that to induce a significant deval uation effect, rats needed to contact the bait in the maze after devaluation treatment in their home. c t.
(9) Effects of Lithium Chloride Induced Reward Devaluation. • Paired. ~. U. ~Cnpaired. r"'I CII:. [J Control. ~. 100. ~. 80. ~. 60. U. Z. CII:. ~ 40. ~. 20. ~. 0. r"'I. 50. '""'. ";. ~1 il. .;: 40. ..... ~ 30. '". '-'. r"'I. 20. ~. ](). r"'I. 0. ...~ z < ~. (. r. :;. ;). s e. ;t. h. l,. BSL. EXTI. RET!. RET2. EXT2. Figure 5. Effects of lithium chloride devaluation on the egocentric task as measured by mean per cent of correct response (top) and the mean time to complete a trial (bottom). Three groups were: the lithium chloride pairing treatment (Paired; n 9), the lithium chloride un-pairing treatment (Un paired; n = 6), and the physiological saline pairing treatment (Control; n = 8). The experimental ses sions include the baseline before the lithium chlo ride devaluation (BSL), two extinction sessions (EXT1 and EXT2), and two retraining sessions (RET1 and RET2). See text for detailed proce dures. Each bar is expressed as the mean ± 1 s.e. m. * indicates significant differences between the lithium chloride paired group and the other two groups .. =. .d. 2,. cages.. ;. Discussion. Ie. 1S. ;)-. le. )ll. ll. he ne. The present study attempts to clarify whether the taxon system is composed of subsystems that involve different learning processes. Following O'Keefe and Nadel's (1978) distinction of the cue-learning and the bodily orientation strategy, this study selected specific tasks to test rats' spa tial behavior learned through these strategies. By. 251. devaluing the incentives used in these tasks, and by changing spatial cues, we found that: (1) rats that learned spatial goals through spatial cues were indeed much affected by cue-removal while spatial behavior was unaffected by cue-removal if the behavior was acquired through bodily ori entation strategy, and (2) the lithium chloride de valuation procedure influenced spatial behavior only ifit was acquired through the bodily orienta tion strategy, but not the cue-learning strategy. Based on the traditional view of learning (Bolles, 1972; !tall, 1943), one may conclude that distinct learning components are involved in the two tasks of the present study. The egocentric task involved S*, but not the cue-learning task. Since lithium chloride devaluation treatment af fected performance of the egocentric task. Fur thermore, from results obtained from environ mental cue manipulation, one may argue that the egocentric task and the cue-learning task were in fluenced by internal and external factors respec tively. Performance in the egocentric task was not affected by environmental stimulus manipulation (i.e. shifting the direction of the starting runway). One may assume that performance of this task was based on internal factors. In contrast, failure to perform in the cue-learning task when sandpa per cue was removed indicates a critical role of this external factor. Given the demonstration (through incentive devaluation) that reward (S*) is a crucial factor in the egocentric task, one may raise the question whether R-S* or S-S* was involved in this type of learning. The answer to this question may be de duced from the results of the probe test in the e gocentric task. Since correct performance rate was unaffected by spatial manipulation in this task, one may deduce that R-S* learning is most likely involved in egocentric task. The role of reward in the cue-learning task may be purely activation. It activates, but dose not determine, the learned response. Since S* is not involved in the cue-task, S-R learning is like ly the process involved in the cue task. Results of the present study are similar to Sage and Knowlton's (2000) findings that the ef.
(10) 252. Jian-You Lin and Ruey-Ming Liao. fect of incentive devaluation depends on the na ture of the learning task. In their report, the win-shift task was affected by incentive devalua tion more than the win-stay task. In our study, e gocentric task was affected by devaluation more than the cue task. In both the win-stay task and the cue task, the connection between reward and spatial cues seems essential for acquiring this type of spatial behavior. The rat makes a correct arm entry guided by a spatial cue to obtain the re ward. This is the classical S-R learning process. On the other hand, the win-shift and the egocen tric task were both affected by the devaluation of incentive value. The behavioral component of learning these two tasks seems related to the ex pectation of the incentives. However, a difference can be found between the egocentric task and the win-shift task. In the win-shift task, the location of the reward is fixed relative to environmental cues. Therefore, it is possible for the rat to locate the reward by environmental cues. While in the e gocentric task, reward is located by specific bodi ly orientation with no reliable external cues to guide the rat to find the bait. One may inferred from this difference that S-S* association is rele vant to the learning process in the win-shift task while R-S* association is more likely for the ego centric task. Results from the present studies demonstrat ed that the lithium chloride devaluation proce dure can be used to investigate the nature of the learning process involved in spatial behavior of radial arm maze. This devaluation procedure was used in the past to investigate the nature of learn ing in operant conditioning in a straight runway (Chen & Amsel, 1980), and bar pressing in an op erant chamber (Colwill and Rescorla,1985). Re sults from these studies are somewhat different from our results in terms of devaluation effect. Chen and Amsel (1980) found that lithium chlo ride devaluation showed its effects in the very first extinction trial as opposed to our study which found devaluation effects only after the first extinction trial. It seems that in our study, in order for the devaluation treatment to be effective the subjects must first recontact the bait in the. maze. There are two possib Ie reasons for this dis crepancy. First of all, from the nature of the be havior involved, in a straight runway (in Chen and Amsel's study) a rat does not have to make any choices, a straight forward approach behavior suffices. Running in a maze involves much more complex behavior which might be more resistant to devaluation treatment. Bolleine and Dickinson (1991) found that it was easier to influence straight approach behavior by devaluation treat ment than operant bar-pressing. They contended that approach behavior was based on S-S* learn ing, but bar-pressing was based on R-S*. S-S* learning may be directly affected by S* whose re ward valuation may be down shift, which is not the case for behavior in R-S* learning. For S* de valuation to influence behavior in R-S* learning, the subject must contact S* once again after de valuation to adjust its sensitivity for the S*. If this speculation is true, then that devaluation can af fect maze running only after the first retraining session, could be understood. Since maze running is also based on R-S* learning. Furthermore, where lithium chloride devalu ation was given may also make a difference. Op erant behavior was also based on R-S* learning; nevertheless, Colwill and Rescorla's (1985) re ported devaluation effects were apparent in their study right after the devaluation treatment. In contrast, the devaluation effects in the present study were observed only after an extinction and a retraining session. This difference may be due to the place where devaluation treatment oc curred. Colwill and Rescorla (1985) injected their rats with lithium chloride in the operant chamber. We treated our rats with lithium chloride in their home cages. When Colwill and Rescorla subse quently tested their rats in the operant chamber, the chamber had already been associated with the devaluation treatment. Rats in the present study were tested in the maze that was not previously associated with devaluation in the beginning of test sessions. Therefore further retraining was re quired to allow the rats to associate the maze with the devalued reward. The effects of lithium chloride devalua.
(11) Effects of Lithium Chloride Induced Reward Devaluation. r. 11. !t. d. e. :-. Ir. r. . Ir ~-. Ie. ,y. y. )f ~-. th. a-. tion is more profound in operant behavior with higher-ratio responding, for instance in compar ing the fixed-ration type with the fixed-interval one (Adams, 1982; Dickinson, Nicholas, & Adams, 1983). The di fferential effects of lithium chloride devaluation on these two behaviors with higher and lower frequencies of responding could be attributed to different numbers of reinforcers acquired. According to this inference, one may question that the profound effects of lithium chloride devaluation observed on the egocentric task rather than the cue task of the present study could be due to the number of rewarding baits be ing acquired before devaluation treatment were different for the subjects in these two tasks. This speculation is not likely to be true for the present study. Because the mean numbers of rewarding baits being a cquired before the lithium chloride treatment for the subjects in the egocentric and the cue task were 56.6 and 60, respectively, and which difference was not statistically significant (p> .05). Thus, the possibili ty of different number of baits or degrees of rewarding experience ac quired before the devaluation treatment can be excluded as a confounding factor to the differen tial effects oflithium chloride on the present two tasks. O'Keefe and Nadel (1978) defined the cue-guidance strategy in terms of the association between a specific cue and a goal. Response learning was not their primary concern. However our study found that S-R learning constituted the behavioral component task. Thus, the cue-guid ance strategy may be the result of S-R learning rather than S (cues) S* (goal) learning. One may account for this distinction by considering the role of bait in this type oflearning. Two potential roles for the reward are: 1) it is an environmental stimulus that may be associated with another en tity (S or R) in the environment, or 2) it is some thing related to reinforcement that reinforces an existing association (Packard et aI, 1989; Mcdon ald & White, 1993). The type oflearning involved in the cue-task depends on which role the reward functions in an experiment. If it functions as an environmental stimulus, S-S* would be learned;. 253. if it acts as a strengthening agent, then S-R would be learned. One of the factors that determine the function of reward has to do with the duration of training. Packard and McGaugh (1996) reported that S-S * association was formed first with hip pocampal functioning in a T-maze learning be havior. However, after additional training, S-S* association gradually shifted to S-R association with caudate nucleus in control. Packard (1999) further reported that the timing of the transition from S-S* to S-R can be affected by changes in neural activities in the hippocampal formation and the caudate nucleus. With neural activities in hippocampal formation increased by injection of NMDA, transition from S-S to S-R would be much more difficult. On the contrary, if injection ofNMDA were made in the caudate nucleus, such transition of learning would be hastened. Based on the assumption that caudate nucle us is responsible for S-R related spatial behavior (Packard & Knowlton, 2002), one can expect that candate lesions should produce different degrees of impairment of spatial behavior at different stages of the cue-task. By using a neurotoxin to produce lesions in candate nucleus, a study from this laboratory found that damage in this area of the caudate nucleus affected performance but not acquisition of the cue task (Lin, 2001). This result reflected that the performance of the cue-task is related to S-R association while acquisition of this behavior is still in the pre-S-R formation stage (Lin, 2001). Sage and Knowlton (2002) also demonstrated that there was a transition of learn ing component in spatial learning based on cue-guidance strategy. They found that devalua tion of incentives by lithium chloride injection affected subjects with less experience in training of the win-stay task, while well trained subjects in the win-stay task were largely unaffected. This result may be understood by assuming that the less experienced subjects were at the stage of S-S* learning, while the more experienced sub jects were already advanced to S-R learning. This transition of learning may explain some of the conflicting reports on cue-guidance learning (Mitchell & Hall, 1988; Packard et al. 1989; Tar.
(12) 254. Jian- You Lin and Ruey-Ming Liao. rard et al. 1984). These discrepancies may be the result that learning components involved S-S * and S-R were not equally affected by lesions of the caudate nucleus. To truly clarify the correspondence between behavior and neural mechanisms, one must first understand the nature of the behavior component involved. Following this principle, the present study used lithium chloride devaluation proce dure and alteration of environmental stimuli to investigate the nature of the learning involved in two spatial tasks. We found that the cue-task with the cue-guidance strategy and the egocentric task with bodily orientation strategy were based on S-R and R-S* learning respectively. The neural mechanisms responsible for these two types of spatial behavior require further investigation be fore more solid conclusion can be drawn.. References Abraham, L., Potegal, M., & Miller, S. (1983). Evidence for caudate nucleus involvement in an egocentric spa tial task: Return from passive transport. Physiologi cal Psychology, 11, 11-17. Adams, C. D. (1982). Variations in the sensitivity of instru mental responding to reinforcer devaluation. Quar terly Journal of Experimental Psychology, 34B, 77-98. Balleine, B., & Dickinson, A. (1991). Instrumental perfor mance following reinforcer devaluation depends up on incentive learning. Quarterly Journal of Experi mental Psychology, 43B, 279-296. Bolles, R. C. (1972). Reinforcement, expectancy, and learn ing. Psychological Review, 79,394-409. Chen, J. S., & Amsel, A. (1980). Recall (versus recognition) of taste and immunization against aversive taste an ticipations based on illness. Science, 209, 831-833. Chinese Psychological Association (\996). Ethical stan dard of psychological professionals. Taipei, Tai wan: Chinese Psychological Association. (in Chi nese) Colwill, R. M., & Rescorla, R. A. (1985). Postconditioning devaluation of a reinforcer affects instrumental re sponding. Journal of Experimental Psychology: An imal Behavior Processes, 11, 120-132. Cook, D., & Kesner, R. P. (1988). Caudate nucleus and mem ory for egocentric localization. Behavioral and Neu ral Biology, 49, 332-343. Dickinson, A. (1997). Bolles's Psychological Syllogism. In M. E. Bouton & M. S. Fanselow (Eds,) Learning, Mo. tivation, and Cognition (pp. 345-367). Washington DC: American Psychology Association. Dickinson, A., Nicholas, D. J., & Adams, C. D. (1983). The effect of the instrumental training contingency on susceptibility to reinforcer devaluation. Quarterly Journal ofExperimental Psychology, 35B, 35-51. Garcia, 1., Kimeldorf, D. G., & Koelling, P. A. (1955). Con ditioned aversion to saccharine resulting from expo sure to gamma radiation. Science, 20, 157-158. Hull, C. L. (1943). Principles olbehavior (pp. 68-83). New York: Appleton-Century-Crofts. Jarrard, L. E. (1983). Selective hippocampal lesions and be havior: Effects of kainic acid lesions on perfor mance of place and cue tasks. Behavioral Neuro science, 97, 873-889. Jarrard, L. E" Okaichi, H., Steward, 0., & Goldschmidt, R. B. (1984). On the role of hippocampal connections in the performance of place and cue tasks: Compar isons with damage to hippocampus. Behavioral Neu roscience, 98, 946-954. Kesner, R. P., Bolland, B. L., & Dakis, M. (1993). Memory for spatial locations, motor responses, and objects: triple dissociation among the hippocampus, caudate nucleus, and extrastriate visual cortex. Experimental Brain Research, 93,462-470. Knowlton, B. J., Shapiro, M. L., & Olton, D. S. (1989). Hip pocampal seizures disrupt working memory perfor mance but not reference memory acquisition. Be havioral Neuroscience, 103, 1144-1147. Lin, J.- Y. (200 I). Investigation of the neurobehavioral mechanisms underlying spatial memory. Unpub lished master thesis, National Cheng- Chi Universi ty, Taipei, Taiwan, Republic of China. Lyford, G. Y., & Jarrard, L. E. (1991). Effects of the compar ative NMDA antagonist CPP on performance of a place and cue radial maze task. Psychobiology, 19, 157-160. McDonald, R. 1., & White, N. M. (1993). A triple dissocia tion of memory systems: Hippocampus, amygdala, and dorsal striatum. Behavioral Neuroscience, 107, 3-22. Mitchell, J. A., & Hall, G. (1988). Caudate-putamen lesions in the rat may impair or potentiate maze learning de pending upon availability of stimulus cues and rele vance of response cues. The Quarterly Journal of Experimental Psychology, 40B, 243-258. Morris, R. G. M., Garrud, P., Rawlins, J. N. P., & O'Keefe, J. (1982). Place navigation impaired in rats with hip pocampallesion. Nature, 297, 681-683. Okaichi, H., & Oshima, Y. (1990). Choice behavior of hip pocampectomized rats in the radial arm maze. Psy chobiology, 18, 416-421. O'Keefe, J., & Nadel, L. (1978). The hippocampus as a cog nitive map (pp. 62-101). Oxford, England: Oxford U.
(13) :. If. J.. ). ). v-. g. ]. Effects of Lithium Chloride Induced Reward Devaluation. ni versity Press. Olton. D. S., & Samuelson, R. J. (1976). Rcmcmbrance of places passed: Spatial memory in rats. Journal of Ex perimental Psychology. Animal Behavior Processes, 2, 97-116. Packard, M. G. (1999). Glutamatc infuscd posttraining into hippocampus or caudate-putamen differentially strengthens place and response learning. Proceed ings of the National Academy of Sciences of the u nited States ofAmerica. 96, 12881-12886. Packard, M. G., Hirsh, R., & White, N. M. (1989). Differen tial effects of fornix and caudate nucleus Icsions on the two radial maze task: Evidence for multiple memory systems. The Journal of Neuroscience, 9, 1465-1472. Packard, M. G., & Knowlton, B. J. (2002). Learning and memory functions of the basal gangl ia. Annual Re view ofNeuroscience. 25, 563-593.. 255. Packard, M. G., & McGaugh, J. L. (1994). Quinpirolc and d-amphetamine administration posttraining en hances memory on spatial and cued discriminations in a water maze. Psychobiology, 22, 54-60. Packard, M. G., & McGaugh, J. L. (1996). Inactivation of hippocampus or caudate nucleus with lidocaine dif ferentially affects expression of place and response learning. Neurobiology of Learning and Memory, 65,65-72. Pisa, M., & Cyr, J. (1990). Regionally selective roles of the rats striatum in modality-specific discrimination learning and forelimb reaching. Behavioural Brain Research, 37, 281-292. Sage, J. R., & Knowlton, B. J. (2000). Effects of US devalu ation on win-stay and win-shift radial maze perfor mance in rats. Behavioral Neuroscience, 114,295 306..
(14) Han-You Lin and Ruey-Ming Liao. 256. Appendix. *~~~~ ,~ Jj ;\l*IJ;%±. Z. * t. st t~Hf ft * s 0', 1£ ., ~t 3M A~ :It ~t -'-t *". ~~~ , 'Bd -'-+ .0,1 'i!li -'-.} :?:: ~' r'b ~R ./11 Aq EC' }j',;! i'r.z i"lJ Ii):, ' 1*- p'lif-; ",(; ;:r; ,3f, :'L laJ 1T Aq I' #~ ~ ~I & ft 1ft ~ ~ 31{ ~ (O'Keefe &. M /.-. CI~. •• ~~MI~.~~~&t m. Nadel' 1978). it 1HJ'- JI /f. '*. IBl 1§ r*J ;ffi:t it $;. *. ~o~**~:t*Si1£,*1§~M~A~:It~ z.t,*I~~&~~~~~~*'~~~~ ~ ~ ~ ~ /\.1~ ~ ~:i! rillf,~ ~ J'ft 11 1!stm~~~~~~:mt~~~*.,*~~~ ~-*.~~m~~~:mtst~~~~~~st~ ~:mt*_z~~~#~~~.A~~~A». -*. NfL:mt 1!. A. ~:mtft~*.o~m~~stft$;£'.*~ *,~~~~~&~«~~.z~ ftz1t~ ~'~~~ ~~MI~~tl~st.~o~ ul;; J+. +~ ;,BII "'-f H, EI1. A J" 3m /.s ~s ~ ~ 3l:il 1:k $- ,!,J., -->- oJ-, !}~ 11c\. WI- i'm pJ:, "'~ 7f':'~ 7L S 1"f "i7)<. Jr-. S f 7R: CI~ A t::I i1£~~.t~#~~~~.~.st~$1!n 8~ Fa' , ffiJ 1£ ~ ~ ~ I~' if ~ 8~ S 0'~ ~IJ ~~~~M~_~z«~~.tl.~~ft$; *mo*.st.~~~~ ~~MI~~tl. ••. *. *. £fL:mt at. /f. ••. ~st~tffiJ.~~'~~.~~. *. •• '~~~. I~' it ~ st 1t $; JJL ' ~p ~ ~ ~ lE 1ii $ .1l1. & Jf ~~~*~~*;~:t'*Si1£.t,*1§~ .stft$;*mM~~~* •• ~~~.~st. ~o~~~~M*,*si1§~.~,*1§~.. *14m 71*. ~ ~ ~ ~ I~' it ~ ( S-R) ~ ffi:. »! -wm. 3M. , :t. t". JJ~ • ~ JM • - ~ (R-S*) 8~ ~tP ~-li. r st. '* »!I. ~ ~p tt, ~Ut t ~ ~I & H' 1ft ~ ~ m3l{~1ii.$;m~/fIBl,*1§r*J;ffi:t~~ff~o ~.M:.f.I·~ft~~·ffA~~·._.I·. $Ut~;\~:lVt·. *S. ~. lli:tj=:*~m:t.e~rf3~rJ~. ( spatial behavior) 1¥}liHY"C. .~'-H~~~OOfi~~~~~.-~~'®~ ~~~~~~~~~H~~~~~.~mR~~~. O'Keefe W Nadel ( 1978) ;fJU~@l~~H~JE §.~n~M~OOfi~*mffi*.:-~m.~~ ( locale system) '~~JH=rm:t*m~ rlOl~:r:\'::f:~:f:th~ffiilt ( place hypothesis) k~H_ r±l:9'}tEfjV:~~ pg1£~1il& ; :!fJ--~;f~§,t~*fC (taxon system) '~~JH=r~ PI {:t(@ ~T~7E § T~S"J~~~j;lli*~= ' NPflJ m~9'*~ijl ( guidance) ,& irs1ili: ~ rPl ( orientation) m fi~~ S"J 5t1t~JH\f.!. 0. t~§,t~ME1i~. 0. flJm*,,'*~ijr~{*t§l@~if'J¥~fJlm. ~~~~~R'~.r±lmlli~~.~fi~&~'~. if:tt:. mRili: ~ rPl ~ ~U ~ ~ {I'm ~1=r ~ f'l"J R ~~ § :flt ~ iL'S"J~rf3~~T_~ (egocentric spatial framework) J-.),. M" (rotation) S"Jn;r:t*1'lHJ! ~{I'm~Ff3~~TPiS"J*tb iL' nJ J;J.~@Rs"J§lUM ' Ei[ , tt~~ , {1~PD : @R'W 0. 1~tEfElg8"J:f:thn l:J.,§fR~'fHiL'ttE:~tt.tl~, 90°. n:J. ~m~~.j;lli'ili~~Rm~rPl~~~ft~X~~ § :flt ~ iL' ~ ~ Ff3~ 1=r m ( egocentric spatial behav ior) n ~ r~~ 1=r ~ J]i$ t~ ji 1'1" t~ ffilj s"J liff Y"C R J'J! O'Keefe W Nadel ( 1978) P!Tl&*Z~*~Jj~Ff3~1=r~i1U1Ci¥J ~~~~~~~.~m~~o.m.*ME~~OOfi ~t&*~[fJJ~iIlli ( hippocampal formation) JjJl'mZF,]lff W'~*~~~~~~~~m~~.~.~~~, JlUtt~Jj~Ff3~1=r~~O~lt~7.¥: (Jarrard, 1983; Jar rard, Okaichi, Steward, & Goldschmidt, 1984; Kesner, Bolland, & Dakis, 1993; Knowlton, Shapiro, & Olton, 1989 ; Morris, Garrud, Rawlins, & O'Keefe, 1982 ; Okaichi & Oshima, 1990 ; Olton & Samuelson, 1976) 0 :f§lt:t.e{fti1'1L**JE~Ff3'1=r~W~ ~iIlli~WWM~' • • ~*~~~OOfi~Wn~~ ( striatum) ~!i:'A~ ( caudate nucleus) ZJt B! ~m f* ~U ~ ~ 3J;r S"J ~ - ¥)i: • '~*R ¥~ ~ Pisa W Cyr ( 1990) RJJ!iilV~f§;f~7U!Y.R~j;llimff~ ( turn discrimination task) ® ~ ~ ~ ff~ fl j;lli if'J¥ 1'1" ~ ( brightness discrimination task) 0 Mitchell W Hall (1988) ¥~~~;f~~~~J:.lS"J*BJJjltE y 7f5j2!S?$~~ m~rPl~~OOfi~kB!~m'ili~tEj2!S?$$~~A -@t~ff; § t~1&Zl!fHii~:;,'* ' ~U1JJ PI~1~J;)E::En rPlj;llim~ff~c~~m.&~*.ff;~~~m~~. z. ue. *. ttl:!: R-S. lW~*~ ride.
(15) Effects of Lithium Chloride Induced Reward Devaluation. RW.tt~~ • • ~~~Z~~fi~,m~ Packard, Hirsh"& White ( 1989) 'Lyford:W Jar rard ( 1991 ) "& Packard ~~ McGaugh ( 1994) tiP ~ l~ [I:iH~ a~ ~if1{ ~tJ:1 % Wr~*~ ~ ~ sl A a~ ~1$1j ~ ~ mom®ft~B"J~*~~.~~~;%ngcTZ~*,~ sl"& IHll~ ~ m *~R~~1$1'J~ PJ§g~i'f/f fRl Z 11' m~~,rm~~~~m~~~WAm~.ff~&~ ~ f'F B"JfjNi'f ffr /f fRl 0 gt~F~M'J~B~~~ r'kJ~rm § , O'Keefe :W Nadel ( 1978 ) :il:fLlf:: t±:\ f~ §;*;% *1CF a~ mmflID ;:]1: ;% ngc 11' ~ 1iiPJE:JjIj , R%SII!~H~t±:\~{ll~~*~~~~~&JJ! B"J~~ (response learning) 'rmr:;,~*,~sl*~:ftfJIJ :lj~!f-'}5E~:i:J£JijU?!H~ § ~l&I'r~~~~*j§ 0 ~rm Sage :W Knowlton ( 2000) tiP~lJtBZJJ!B~~~tEl5~*~'R*,~ sl * ~ Z ~ 1$11' ~ win-stay 1'F * cp J1H5} i1tfi ~m jIjfj. m. c ~. ~. I ~. it. IJ . P. ~. iIll ~. ~J. ~. v-. B"J. 1']. ilJa]. u. 4;. 1,. & &. $. e. I{*. :yr. ••. ••. ~12SI~Mz ~~'m #~~83~.~W~ .tt~~~~~ffi~p)G~%Stt.~&JJ!rm~~~~ ~i B1+1 • ~ a~ ~ ~ {Ii {oo: (Garcia, Kimeldorf, & Koelling, 1955) 0 Colwill :W Rescorla ( 1985) flJ m~@~~~~1tW~~U.mMWfi~Z.~r'kJ. ~'~*~mOOiltZmw*~Wr.~ • • *~~1t ~MB"J.~rm~~'.~~ • • ~~mWff~®~ ~cp~~s*®ffi~p)Gft°~*B"Jm~~~mm• • .*I!#;~'I1t{~{oo:a~.~~,~j;JIE~~5tR (positive. reinforcement) l3tfj~.:1::a~f#f'F~i!JU%s1'J~B~~~ r'kJ1rB (Chen & Amsel, 1980; Dickinson, 1997) ,{§. *~~*~mtEUIE~~~~@B~®~OOff~'. (radial arm maze) 1'J~ IlfE Sage :W Knowlton ( 2000) flJ m• •* I!#; ~i 1t{Ii fiB: B~. ~ ~ mit ~ ~ %J ~ 1$ $ win-shift "& win-stay ~.i'F*B~~~~~~*j§lf~A ' M:fH'F*~*. {JU~D. :. ,&;~jllil~jfj~?$. 'Ill. 0. }\. '6'. ~Q*1f*~*'~SI*~Jt.{)Ji'f&~B~~~. ~F~U. O'Keefe:W Nadel p!T.:1::~~a~Ri'fl~:l~JijUl%ZF",a~. ~~~.~OO~~~*~5E~*~~~'rm~~~. .~'rm~~~~m~."&fi~&JJ!OOB"J.~~ ~omu.~~~@~~B"J.~~ff~~~r'kJ~B"J 1iii;14-m1~ iilfi=±3Z,. mnh7U F!IOI:;-.e;, *' -;:r;:;.b.~ I" ~ EH'I'fi JmJ L=t:.IIU S ' 'l.mPJw<·"I-": ~YLdl'J =:L.11':I'J IJ ~..!:iI:. --- l' FJl:::irA ~ 'WJ..R -.1::.. *f!1J ~ ~ * ~ ~ {1( iaH7'R *' ~ sl * ~ 5E{ll 14:ilIj:{ll i1[ ;l:t• • ~ft~=OO~~'~~~~~~Wrm.~* ff~,m~~ff • • *I!#;~ • • ~B"J.~,~~m. z®~~m~~Zo**~~ff~~~pq~®jfjN ''''Jt-. ml!#;mm~~~.~~M~~~B"Jfi~~mM~~ ~~.o;l:t~*~~ • • *I!#;~.~MB"J.~e~. ~SI"&m{ll~~m.*~pJT~p)GZ1'J~&JJ!~~~ fi~fRlMtE®.~~~~~,u.m~~;%~~~. :lWjJD~~7tp)G. ,. Jl:[. f7G =-+ iilfi =± ~ Y.i::. "'h ,:Ai FlIOl :;- .e;, -hb Ifb ::Et i-k tl'l:! Y.B FJi; ~~ ~ 1-*p'J !17RaJc.~7R ilYLr::l'j--=r:::.IFlJ IT m ' 'lx.g"8~ ___ " ·1J.-\~;f\71<*. m-.j{{'I1:Z~F,,'1'J~;%ngc. 0. win-shift {'F*a~s~r",' , rm/f~.7Gp)G win-stay {'F*a~7Gp)Ga~F,,~ , {§.~~~mmf.l{'F*B"J &JJ!IE~$*~~.o83~~~'~.~*~~. 0. ~ :tH~ ~ lJjlfB :*: fJz:~~ 11' ~. Z rJg iifJ 7t ~ s-R '. o;l:t$ S {i;~f~:i:i:JijU. S-S* Ez R-S* =.~9fi*j§lfJA lfI: ' R {i;£H'J~&JJ! ' S* flIHi;~~J+I.~~{'I{LtlJ/f*a~ ~*W~iJtRJIH.§1. ( reinforcement) mlUJa~JijU. S-R 0. fe. 257. win-shift {'F*a~~~ r'kJM:iZ~m£1 ' 6Jz:lIt1'J~JlJ'N S* 1=i ~m a~ ~ij~ *j§ ; wi n-stay {'F * I2SI i~nHl'I'I. ~ {Ii {oo: m; A :!3;~~ r'kJr@:i , 69:. S-R ~~~~*j§ Sage:W Knowlton (2000) a~Wf~*j§*.~7flJm • •*I!#;B1H.ili{00:. .~~~~~~W~~m$ffi~~®ff~&JJ!~*. B"J.~*7t~~OOff~B"J~~r'kJ~~0fiB"J,mR. iJJi'I"J~~83~:i:~JijU.PJTsl ~. ~5E~J+I.~7;)B~ffi~§g1J. ,. @~1'J~&ff!~±B~iJtR. 1t{*~f[!lJilt*~a~~l~**,~[@1J ( drive ) */J\~~ (Hull, 1943) S-S* ~ R-S* B~~~ r'kJ~flIJ~W¥U fflJtJj ( expectancy) iJ\]~~~ , S-S * ~ ~~J~JijU i%zffiJtJj ~.Wt±:\lJta~§g1J ' rm R-S* ~~~~1U~E!~B~!f-'}IE ffm&JJ!ffi~~.Wt±:\mB"J~1J,mm.ffm~~ 0. i¥JiJtR1t:l$])'t¥U S* pfr ~i'fa~~l2SIiliiOO: (incentive val ue) ~Il ( Bolles, 1972) "l2SIrtt{lIDmi-£~m;;,*;%ngc~ OOfim$ffi~p)GB"J~~.~~~~@~F~~OO~ H~aB"Jm~*~U.~:-'M.W~M~~~ .@iltff~BZ~B"JSN?=,ntE$m~.~~~. flmm1'J~&JJ! ~±Z {1(tJi. ?. ~m-~M~.~.W~M~~~.@iltff~. Z1'J~r'kJ~. ,. ~1/;:~. ( S* ) ~~~ win-shift ~* S-S* tm~ R-S* fllJf;fU*f)t~. '"~. BE. ... ~~ffi.Z~~=tE~~.~m~.~~~u ~~OOmfim&JJ!B"J~tJi'U~mOOiltZ~~ff~ ~15~-.~~.~o~~~ *I!#;~.~M. ~~.Z~OOfi~':E~~~m~.%*~.~fi. :95Y'i-- ' '6 iIID ~ ~ F,,' 11' E&1~ ~ ¥U jj.* I!#; ~fH • iii {oo: fj rr; B~ ~ • ' f!1j~~-*_~. S-R ~Y~*ij§Jf;;~Z~F"'1'Jffi& ' E&Z~~. , flU. S-S* ; &Z ' fllJ~ R-S*. 0. ~N~i~Y'i--tE~mJijU.p!Tsl ~ 0 $7~m~~;%ngc~OOff~~~~~@/ffRl~. ~.~~~m~Mz*~;%ngc,*m~.~~m. ~ 11' 1: ,~~ $:JW m ~ f1UIlHH lliIAjfj~~~~ff~~*~~*,~~~*. & J! 3Eri1t ? 83 ~ S-R lliI B~ ~ ~ ~ r9: 1'] ffi ftJj Il\] ;fB"t ~ tE ;!tr:p , jffi S-S* "& R-S * lliI n\]~9~*j!j~~ 21 trf][ftJjB~m;. * Jiji] ?%z"& ~I'I. ~~ i~ too: *. [all. Ie. ~'~~~*~.S*®~~.:1::~~~ff~&~o83 JttOJtt~1fB ' :E~{l£@fH.ili{OO: ' ~*~.¥UJ;J S-S* "&. ( cue-learning task) "& § ~ cp JL' {'F egocentric task) ,;l:tft5jU{i;~*~~*,~sl~"&fI{ll~~~n~~~~t. JA. :;15. 11&Mfa. un. .~. ~-'f. R-S*. ~~~~~ r'kJr@:ia~1'J~~lJt. z 1'Jw~lJt. 0. ride devaluation). '. rm/f~.J:.J. jjlll~t*I!#;~#i1t{Ii{OO:. S-R ( lithium chlo. a~f'fl.)-f~-{t~ mJj~~~{l£~'I1tW. 0. *(. ;%ngc~OOfim°tE~*~~~*cp'R~-~5E~ $(tE*m~$w@m):W~.~~~°tE~W* $ ~ 1/;: ~H 1t ~ ffi[ ~F:tE ~:i:i: cp iIID [nJ IE{ll ~ E.:W iIID. *. m.
(16) 258. Jian-You Lin and Ruey-Ming Liao. lli~~~~~~%'~~~~~~m~lli~M~~' JlhmIil&Ji:f-J~~ (Lyford & Jarrard, 1991 ; Okaichi & Oshima, 1990) RZ ' tE §fJtCP,Cdf*CP , R [email protected]~~~.~OO~~~.~~~'ffW~m WU {~~ ~ *.fD ~HI. t?iJ gc ft t±l:fJ! ' ESIll:t)t ~ R ~ {f\( ill: 1m' 1.iL ~ Mm 1iI&~ !(!iI:f-J;!3; ~ F,,':f-J ffl ( Cook & Kesner, 1988 ; Pisa & Cyr, 1990) fJt1F'~t1HilIltzDT : ~1rm .fifflR~~~ • • ~~~.~M~.**~~M .~8"J~~~1~i¥J~1i ' flU~fI~F,,':f-Jffl1'F*J1~ S-S* ~ R-S* Z~~P*R:@J:.t ' R,ZflUffl S-R ~J:.t ~ ~-*m~'~~~~.~~~~~Ii~~tE~~ 1$11' * cP Z :f-J ffl R ~ , flU ~ fl1'F * a"J ~ F,,':f-J ffl !Ii S-S* ~~1i*2i~J:.t ; ~~l~IH&~Ii~jL'= ' flUffl R-S* ~ J:.t 0 ~~J:.Jili.* ' OJfj~1~:taU~*~ ij I .1.iL~ ~~mlil&mm~Z~OOfiffli¥J~~~~~Ji-*i¥J 0. 0. 0. »... ~~~. ~~~~ fF~ :f-Jffl wll**~Mj7;j7j7d\'fb3@~[l"& ~,%~e~~tt~~.Z~§Jt~~~&~~m® ~1$CP'~-~.~§~~~li~.o~~~~~ ~'~Jifi.*~m{'F*~~o.&~tE~@.~ 0. zMffg~~j7;j7j@lEPJm{'Fffl~.§Jt~~mlEPJm,~. ~jlI}~PZillA nj7;j7j0~:tllitR~~6~1ifHJ~. (llil5m CwlOO. a"J~ 5*11. iij>*J~ , * 28 051 ' Jl 7.5 051 ) #1'F*! *o~-*.~Raj7;j7j~~.*~lEPJill*~~~ HJ5~~ttm,®~~~@~m&~~m~*M. ~ ~ jlI} lEEH11! (j"J i~Vr)( :1J J:.t 1* 1rlll11H~ (p seudo ran dom) ~JE ' j;),j7;j7j(IEINPZili~;j:§~~ffllJJl:flU ' ~~~9!. m. 0. ~~~i¥JlEPJ~y~~ttfi»,,@~o~~~fi-@.. §Jt'~~*~.~OO~~·~~~§Jt~~~JE~:1J ~*~Affl~~~A~~~~.~~~~ZA~~ 1$CP*~~o~~~A~1$cp*~~li8~' • •. 0. ~m~~~.~.800,y.~~§E8a~1$cp~. ~8it~. W,~~~1*m)\~*a"J;tttt Wi star o"b **s.!fijl , §~~tJ~*~g~~ff,~iJJ~cp{}~~ l~i& ' ~p tE ~ ~ iEJ: ii:1 * ~ ~:tJl!!, C,d:J. ff,~ '¥: iJJtm cP yemJ:.t~ • • n.m~.o • • '¥:&~~m~~ !3V$J.t~tEJI~ 23 ± 1 N ' if1<zii§~~A::ffl12/12 !J\~ (!f.J:. 7:30 ¥UIl§EJ:. 7:30 fflSif) 'i-J~m~~a"J a'1FFl'flUIBI~tETq:-~li¥UT q: 1iJ!!ti * s .!fijltEiOJ. ~icp)fG~-=y§ E8uz:tt&iX*:#:§i-~~ , ~iit¥}tljm 0. m. ~-~i¥J~OO~WJifitt~~~. tt~~w~~ • • OO~~fi~'*S.!fijlmm~ffl§E8~tt~Z. m~ 85 % ' ~fltt~fIJW ( ~ B f(-] 14 :R:tttm*4 ) m~~mm-~.M~~@ • • M*c*~~~~ t?iJ ~ §Jt Z 1Ee ~ ~~U+ ' ±&J 51ii§ cP IH&I ,C,d:J. ~ ~ pfT $[ IE. m. IIC,d:J.~~*A~([email protected]'fflUJ CP~~~.,~mIEZ f!:j~m1~171J 0. fjf.m),~1J8itfi~ *~*~~& §fJtCPIc'\~{'F*PfT ~mi¥J~.~.~~~~EM~A~~1$'~~~ • • '¥:~-.~mooZcp*o~1$~tE.~800~ ~n.J:.'~W:$*~-~~300~~KA~ID~ 0. .,~~cp*~.ffi~m~~A~gEM~m~~ lEPJili ( j7;j7jj7;j7j;j:I3JfE 45 0 ) 0 ~-@~Blilia"J*!J\* 61. , Jl7.5 0~ , ~ 15 0~ , tE~@~Blm~*ilffij ~~-@m~~1$~~~~ttm~tt~mo.&m ~ma"JttiI}~*!BW~Jj*JLZ8J]1i:1 (Meiji) Jl!4lJ5~~ , 0~. ~-~Jjf(-]m. 350 ~U 400 ,,€;:R:. ~ttP,~~m@ttP~~~Yli~.ffl~~M* ~*.§Jt W.~~.)t~~A~1$~mi¥J~~& %~~1${'F~ffi~~~OO~~omm~~{'F~~~. *~~.~~*s.!fijl~~%~{'F*~~±&JlEPJili~A ~ (number of entries ) 1j£~~~~=* ~~§Jt (n = 31 ) ~~ 27 ~§JII**3i¥U~1~1~* fit ' J:lP)~:f-J &~1~~1~IJ§Jt (response probe test) 0. 0. ' 51JjUtE~~~Ji1-J ' IZ9 ffl~.~.~ttPM.~i¥J~.,mYj7;j7j*~.~ ±&JtE~1UI}~m~ (extinction) ;113';l:JlCP~1-J ~* .~z-~*~*8~.§Jt (no cue trial) ,~-@ ~.~~m~.§Jt~{'FM~.~~~~~*o~. *8~.~cp'~1$cptt&~m~@~'~&~~ ~tti:E}; ~%~§JtjjJ:~.R:1J~tl~li~{& ' lmnOO ~ • • ~§~~~~1$'~~~JiA=@lEPJm~~ ~81±l~t$ , ll:t.~A~P*2i* ~JNPZi~JiAJj(~§c ~{rmm~~§f3j7;j7j;;Xa"J.§A. 0. 0. .~,. • • ~~*~§Jtcp~~~~.~§Jt~m~w:. cp*~tJZ~OO(®~~AlEPJili~~~OO) o~~. .*8~.~z~'~~~fi=~~tt~~.*~ ~1'F*a"J¥}§JII** ( retraining) ,~§Jt3i¥iJ~~a"Jm m~If~*B~fi.~~m~.§Jt,m~.~ffi. *8~.§Jtz~~~tE~M~~~1$cpm~~~@ ~.*o@~~~nJ:.t~n.~~~@lEPJili'Y~ ;j:§~~~J)j{flU 0 ~~~A~A~ {rmlEPJilii~flUM4~8t±l '. ~~o~~M.~M~.*~mz.~~ • •. ll:t.§Jt~P*2i* ' J3;§c~!I<nJ:.t~*''R*8~.~~ ~mBmm.*~~ff*z~~, • • ~~~. ( lithium chloride) Qjlj § Sigma 1t ~ 0 RJ ( St. Louis, Missouri, USA) ~.lll]li.lN1*fljm±'3'l!l.~1li! *~~~ffl 0.15 M ' J3;~*ff~1£Ua0JTRmtt~1 10 ~:ftmzRffl$?lM~If (~P 10 mll kg). ~1-J~Hi~{&~~PtE~t?iJmz~OJ.njg:f-J~E!1Ili!~~ M.~~~.*o~§Jtn.~ffl=m(~n=8) ~-*JlJ\'fb~E!1lE1gcfHJ3. (paired) 'll:t*J3.)t~AtEUZ:7'Gf(-] 3~~J5~~~15~.'~-=Y.lEIg~ffMo~j7;j7j. (mg). 0. 0. {*. 0. 0. :.
(17) Effects of Lithium Chloride Induced Reward Devaluation. 259. ~. m~~~~~~m:~$-mm~m*_*~~m ( control) , If:tffJl§t~J\:j±Jll[g?t.alHH''Jff~11WF.S':tm *_* ; rm g-ffJl,§jilH_:7FJicfHJl (unpaired) , If:t m§t~~§t ffM~~&~W~*0~h~~'. :iiS"J~PJm~wil*'*jli'&~f)H13.1T8"J~m ( {JJ~ " -1- " ~~~) 'rm lct~pfT~13.lTs"J~~*WII*,*~pfTPJ ~1'J. i"J. !:lnl!;._tt~1B~r~~ff.kJ1±§t~OZ:7t:ClJ~h1tt. 20 /N~ ~~~~m~oo~~m~~~,oom~m~ • •ff. §Je\ '. ® iJ\: "E )0 ~. ••. 0. M~~*'~~*~ffM~.~._*~~~*o R-*~~.~~~~-~~ff-*,~.§t~~ :j±§t~:ii*J7~h1tt 15 ?Hi~ 20 IJ\~ , *2r ~ • • ~':tm*.*ttMo~._W0~h~~~. ffJlJ3U '. * ~1L;J( 13't ' • _@cfH§. S"J §t ~ mJ"j~UJHf:\ :tE ~P] lUi ~'R. ~~~*0~h~&~,*~~.m~~.~oo~. ~1~1J1H~{~. , fln~1'J &H!~~miJ~i\. 0. ll:tmU~\fj~~D. 1W ~ z t§ r#J li'JiI **fj tf ' r'i - S"J ~)j1J a~ §t gJ\ pfT ~g 1'J. *. ~Z~mo.~~~R:iiff-:;R,::tt~1L*.~-g: ~c~~:nJ:~P1WIl~~fj~pfT~ gtlle~1~ ~:J3G${J\f'P*~§t~ , ~1j~iJJ~m. ~~.R~ff.-*~~.s~m~,~w.Wft 9, unpaired: n = 6, control: n = 8)~WM~~*,~~~~$~fi_.~1fffiPlo* ~~.~~~~.~ff~~~~~:;R~ff'~:;R~. ffJl:n~ (paired: n =. fi1L*-g:~)~-&.I2]:;R~ffm~.~'~=& ~::--:~~~wll*,*-g:~ , .~ffifj~&~c~~:n~P1 ~~. A.. Jlt~*3ZJt 0 ~ • • *~~.M~~~~~,.~~a~~ m~~~~~~~~ffm~~0~ff~~~-~ft ~12]1[Q]~~ , ?tJ3Ua~~7t~ 0 ~-Edjl~I2]:;R:ii1'J m~-g:~ , rm~=&~=:;RJIH'J~wll*,*~~ m~ 'i~!:ln::f:tt~*~:ijF!J~lij;*f\: ' ~§JilA~~~~lja*~~*,~PJ ili*!IiM*~7Jf~ OO[~ *~:ij ~ 1f s"J~i3~~:n~ IQ] Mrmwil .~~ffi~,~§t~~~-g:~~~~ftl2]m~~~ .:~-~~MMm~~.rr~~~~~A~-m .*,~mM'~=~~OOMmm~~~~-~~~ .*,~mAD~~~!IiM*~MM'm=~~MMm. ~. ~.-m~~~*,~m~!IiM*~M~m=m~~~ *,~mAD.'~~~~MOOm~§t~~.=m~. Mi*. Z9. ~.*,~mAD.~~®!IiM*~m~~~c ElfXcp/~\1'F~ 1~17W:;R~3@,~~t§'t~ ( P1*~*,~. ~-(~~~lE~)¥m.~~~~*z&H! ~.~~~*,§t~~.*,~~W~$,~~m~. ~f'F*Zwil~fj~pfT~ ) itt ' .-=:;RmU~§t~". 1J~~~~o§t~~~.~~m • • ~~$'R~. ~~$*~~~~~.~~~~a.&'l~~m~mm l1tU~~z~~i;m~r~~ , F ( 1, 30) = 16.87 ' p < 0.001 0 ll:t*a*.ff;~§Je\a*~~~~~{'F*fj~8"JWII*'* ~,lit.~~~@~~*,~lElit~m*Pffi~~.. [1. '! g. El. 1] ~ ~~ :~ ~ /?(. ;!!!i. "". lJI:. 1Z. 0. mA~~~$~l2]m~m,g~@~mm~~h~. ~ T\[. ~'§'IT ~ AD' f~ Z ~ I J ~ ~ (f ( plus maze) c ~~~~U~::tt~-= f[8j-g:~ , -g:§Je\~~~;tG~;ft ~~~~a~~~m*~'~zOO~~~$*,a. m. " tt. ~ ~. ~ iji. m +. *. ~~.~~~m:iiA~-~m~,ft~M§t~~m ~~a~~~~~~~.Ro~.§t~.~m~~ ili.ffi.:n~~~.,~~:n~~~~!IiM~§t~. .oo:n~m~o§t~~~~~~~OO~~~§t~~ cpJ~\{'F~ilJll*i:¥1. '. am;;Jz~~~~ilB~. ,. &i1~~J1)Z-{[Q]. ll§ili{'F~m~~PJm ' jU~~IID~:j±m~il~PJmS"J~Ff~ W1JrPJ 90 0 s"JNPJJ][¥HIfflOO[#-!1!~0~h 0 .,~1f#?f. *~.oo[A~M~J][*~'&i~~~~~~~~'. f. ~~~.~~~~m~m:iiA~-@~m~,~~. 4. fi~~~m*!IiM'~M~~~m~~~~Jt~~o mU~@~M.~R~m§t~~m~~J][~m~~ g-~J][~!IiM~~M'~::f~m~*pz~Mo~ ~~~As"JNPJm~~~:ijNPJm ' ~IJll[2~~mlElit&:H! ; ocZ ' 'E~A*OO[~:ij~m~lj~c~~tf~&~ 0 ~:7::::!*. ;Z. ~. 8. qij. 0. nvt§t7J''l'fT. 0. :[Q]. !?f k\. $JlA'F~li'JiI*i:¥1fjtfpff~. ~1i*wll*,*-g:~J\. If:tf'F~S''J~1~1J~H~~5!!. = =;R S"J 4i B lEiilf$ (lElit&~!;;Jzli)( / 5 ) 1± 80 % J;)~ " 1f§t~J\: (n = 36) *~!1f 11 *s"J~JII~5i¥Ulf:t{'F~. '. ji;jq5t:frr~m FA Borland International Inc.. pffliff. ~z~ijUfE~®::m Statistica ' ~1'JZ~,~~{.EBii'~. ~~l2§r.~li)(fttrr ' .&'ll2§r~J!'t')(ft:frr& Scheffe • •tt.' •• $~m§t~~fi~~ma~~.~ m1~~~=&:f:f~jHfs"J~-'<J;)p < 0.05 ~$. , rm~Fifljm~l$Ji-s"J$;tjH~'R*, [Z;Jll:t ' §t ~ a 1i'NM~ *'~~ ~ 1ttflP ~ * liff *U * ~ fll. 00[ s"J ji ~Jt ' J;)~J:1mma~1$: $ *~~ S"JH~r~~:!1't)]D *~ ~ • 1f~ • • ,§t~.~.~~m-m~m~A'ffi S"J1'J~ffi'~. 0. •. ~~~mmW~a$*~~~~m~J][AD.~~. *,0. ••. ~=(~~~lE~)¥m *~~.M~~ .~~~.*,~~~~~~.~*o~~~=m~ ~a~~~mm$~~~~m:iiAR'=l2§rmii' ~~~.~*.ff;=mM,~~mm~&~~~m 8"J~,~~*;li~j~*~ (p > 0.05) 0 T[iJm1§.ffJl§t. ~~~~.~*7t~~M'P1.~fi=~rmii'~ ~~.'~*.ff;-=mOO,.~mm~&~li~m ~~,I~~P3J~.~*~ ; F ( 2, 21) = 3.88 ' p < 0.05 ' F ( 4,84) = 16.05 ' P < 0.001 & F ( 8, 84) = 3.03 ' p < 0,01 5-:l~~mu§Je\'IW;tJ!z~~~~jG~B~ r~~~1'JM.&'l±~~Y:*~. ' R~~-;Jz~~wll**-g: 0. ~ (RET! ) $~?-Hrr*a*.ff;=ffMr~~S"J1£~5i¥U. ~*$ , F (2,21) = 5.88 ' p < 0.01 ' Jt'2mu~ B.
(18) Jian-You Lin and Rucy-Ming Liao. 260. 9::1I'1"J=:*,gFR1~Jj'Jl):~*ii;lt,l'j~*¥f; (p > 0.05 ). Ji~. *it~~;j;:f1}~)llif;*~~J\Z 2fJ:!;j7'G~s~r,,~JilJ Scheffe. , *15*~J5!.Jij1lg\\1l~C!tt%r:tZ 2fJ:!;jjCf&:S~rR1);{l'j ~~It:±J][it~*~C!tt*,g , F (2,21 ) = 10.52 ' p < 0.05 ' ;!3;t~.Jillii\bk~citmB"J~,~~f1ljmillJjJFi!lf*1'F ' F (2,21 ) = 6.70 ' p = 0.055 b\:[)=B"JjU4Hi'djDJ! =: *,g 5tilJt:tElil ~ Tb\:[) CP~~*-f1}a)llif~~aJ\cp 1't~c~;J'<S~Ht ${&.lt~. Z~OO~~,~It:~.*~~.~~~.*R~. 5t~:tE~~~BZ2fJ:!;jjC~~~·~~~~~~~ [TCj ilID S~ PQ: ~ 13 Ij Ji1]l'IltsJ ~ ~ 1:1£ ~ fJT • *5* 1-*] :iF ff:tE~=S~pQ:B"J*r:trR~~J1:ii:fIjI*]~*¥f; , F ( 2, 21 ) = 5.24 . P < 0.05 ' 13=~~ , ~ , [gS~P~FlIJ*ii !:*]~Jjd:\! (p > 0.05) ill;~ B"J Scheffe $i< ~ .~ JJHIIil ~C !tt%r:t 13= ~t< r~ ~i,i: pfT :{ ~ B"J S~ rRUJFi!lf17 n5: :~.t:J[J.l.i!t.*lIlciHs F ( 2,21 ) = 9.99 ' p < 0.05 ~. u~~*m~~g~~~.~~~~Wtt~#~. S't ~ Ji11'*,,'R ~ ~ ~ n"JifJt $ . ~~ 5t ~J\ 1)) ¢r toN :l~ n'b *l~. .~~m0~~nm~g*~~a~~~~*w# JWjJn5t~jCf&:*,,'R~~~1'F*B"JS~rr1~ , ~tX9:*i)Jjl ~g~~mS't~Rff~~*CP~=:~~(~~~~ Ji:N. flj ~ 1J ) pfT:{ UI"J S~ r,,1 :R: :!:.t:J[J.lit ~ * lIlc iHs). §It:. It:. E8 JJt PI 9;n 1f1 g. ~ ~ @!H B: 111 {[!'i: B"J ~ *. R Jjf~ Ii:flj 5t ~J\ it. .~~~~*cp~Jiit0~~&~'~~~~Ii~ *,,'R ~ ij I~ S't HJ\ (I"J ~ r,,11 j" ~ :IE lit;;~ =t~ &: JJ! Ii] [J] (~jlf 00 If. :st ) ~§:!)~ § fit cP {} 11' * (I"J &: IJ! {iR~mlj~J\*15*'. r-b\:[)~S't~itrfUf~jpUi~~iliJ~n"Jmlj~. m~Z:IE~$*m·~*~m1]~~¥f;.W~aJ\m tlB"J.lt~*3i*]~~~'lt (p > 0.05 ) TIRl~5t~ jGJ-&: ~i[glif~J\Z /1" ~s~ r,,~ , rnJ t~ill;11' 1fl [j(J ~ f§ 1iR *fE. ~~~'~*~~fl~~¥f;.W~~m~ztt~~ ml~~~ . F ( I, 35) = 4.61 . P < 0.05 .. JJt J fU:i5* &: B9( te S't ~J\:tE § fit cP ICo, 11'* cP lfEJ ~ ~ (fj:IE lit ~ tt ~ WI [j(J ~iliJ ~.~ Pfj ~ph~pfT j§ ~ loR ~WIJ 1!~ B"J 29: ~ rruff pfT T ~1f:. . ~RW~S't~J\~f~ §fltcpICo,1'F*i& .. ~&:JJ!~OOg:R:,@~.ru~m.p;:IE~~ffiita. .',. ~li( • • MK:st)~m • • *~~.~~.. W~It:§flGCPJCo'1'F*B"Ji5.~;l(* ' ~Jji*15*~5jU:tE -:J\fj"~~$.'&19-=:J\B"Jm~1f~~~wll**~~ ~ • • Jifi:J\~~mo~~~=:m~~:tE19-.~m mcp~2f~K~$~m,=~~m~~ff~.~* .~=:moozm.u'&.~m~z~.m~.~* $ , F ( 2, 20) = 4.54 ,& F ( 4, 80) = 2.66 ' ~~ P. < 0.05 ' rm~1L1'Fm~~,~JlU*m.lf*¥f; (p > 0.05). 0. ~,~. *iil-*J!~*¥f; (p > 0.05). :!B~19-mu~J\'lfI3'tj;ZKlit$Jif1'M¥~~~;l(. ~~7j7!tt~~;:j;:¥}~)llt~. ~~J\tHr:tJi11' Scheffe$t&.tt~ , ~~J~.JiI.@c~*,g~ :IElii$SJj.1glt:itRlH~#JIj~:r:t . F ( 2,20) = 9.01 . P. < 0.05. TIiill~5t~J\itt§.1~1j~J\'I'llJ:ff!cp/f~~jC~~i[gl~. ~M~~~~OO'~~~~&~~~.~*~:iF~ ~J\'lfI3'tft'&)(1i1'FmZ~.~ii~l'j~*¥f; . F (4,80) = 4.64 ' P < 0.01 "& F ( 8,80) = 4.32 ' P < 0.001 . ITO C:::C:*r:trCl~Z~,~FlU*iil-*]~*¥f; (p > 0.05 ). 7~1j~J\'I'fllf~,Z/I"t8S~rR1Ji1}M"-. 4.63 ' ~~ P < 0.05 ~~)II**~~cp (p > 0.05 ). .. 5.L®tt§.. ~~'Y..*~,~,. ~ J51, 13=~ =;);: f1} il)11 ~;5R~ 8J\'& ~ ~ *~iil-*]~7j(¥f; , F ( 2, 20) = 3.57. *15*. m~. ~A Z ~.~ &F. ( 2, 20). =. 1£~~~B"Jt~~~~&~~;;jz. =:*r:trp~~.,z~,~J:!;j*iil-*]~*~. Jff~*it~=;:j;:f1}a)II**~~J\&~=~. m~~HJ\CPB"Jt§.~:r:tJ!:11' Scheffe${&.tt~ , R;ti*~9'J~ it ~ ~,;Xf1}a}11 ~~. ~,l)'"& ~ =;j;: t~ ~ ~ ilJt cP . ~I.. @c!tHr:ttllJ:ttR8j~#JIj*r:t~~.~ .• ii~l'j~*1'F . F ( 2, 20) = 7.13 & F ( 2,20) = 10.81 . J:!;j~ P < 0.05. lli~~*ml~~~~~M~~~~~Wml~~~~ ~ ~J\ 1£ § fit cP ICo, 11' * cP R"J ~;Ey! ' :1=t &: JJ!lf.llt ~ T ~ttfijC~~~.~M~~~OOlli~:R:°.~Z' i[3] irs! § fit cP ICo, J:\:? rp~ 11' ~ Z ~ ± ~!f !!t # 5t ~J+I. ttD ill ~~~rm~.·~jC~fl~~~.~~T~,p;fi. •. ~~!!t#MZ~~. ~M¥~~~*~0mm~~. ~~.~*~~~:tE~wcpm~~m~0~~~~ tJj J51, , *~5t~J\gn5: § fit cP ICo'1'F*'I;l3ff!Tf1};:j;:*~,.. ~~B:ttD~~u~~*~~~gM~~~ka~~ {1I1[l'l:T~1f:Z~'Y..*'$~.¥~wB"J1'F*cp. P;itW11'~. B"J&:~tt~1¥&~1jH~J'f-3'f-wUl~B"J~~~ 53J'f-' f§~ 1i5: 11' ~ #; if *,,'R . S't~J\:tE i~IJ ~J\ '1';l3 if! cP ~ :{ ~ ~ 1)7 S~ r,,~ jC ~~~.~~~**~'ml~lli~&:~~.~~M ~wm~~~~Z&~~~ff¥~m~U~~~ *'~~~~fftt~~~Mfl~Z~w~m&~U (j"J~g1J. , %15 *~;EJ! 13= ~ =;);: f1} ti)11 ** ~ aJ\ cP =: *,g FR1 B"J ' F ( 2, 20) = 4.54 . P < 0.05 ' it~=~7~~~~j\ cP =: *r:trR~ B"J~.1~~ilil-*J! ~*¥f; , F ( 2, 20) = 3.36 . P = 0.055 crmtt~~;);: B"J7Pl ~~~,&~~;);: f1} il)II*5R~ ~ cP , =: %fl r,,~£J1:J:!;j *. :IEllt$~J1:m:futj]~*1'F. ~t gifli *m~~~~~*~~~h~~~ff~~~~. ' flU' O'Keefe W Nadel ( 1978) ~ It:*JR~~ij 1"& tHHl[~ rPl*l6fin"J5E~ . jJt~ te*,,'R~~ ~ 11' *"& § fit cP ICo, 11' * 'i' 11'1T ~ 1~1J ~J\ (I"J BlMm f~ J:\: ' ffi E8.Ji1 ~ * ~ 1[1+1. tm 1~ 1[l'l: & 2.\<: .ffUf!~IJ i~ (j"J Jf,~. w,~m:-,.~~~~*~S't~:tE~~CPB"JR ~rnJ~l][t~mIJ. .~~~ • • ~'~~~®~.~8~WI~~flm *JJ!S't:j:J! ; rm §flGCPJCo'1'F*ZS't~B"Jfj"~~JJ!~U~ ~~~~.R.~.WrmS'tm;=' • • *~M. .~fi~~~ita~~.~B"J.*RWI~~S't~:tE §flGCP'[}1'F*B"JfT~~JJ!~m ~~~1'F*~fj"~~JJ! 0. =. '. rm::fi5.S't~:tEtf. ~1t:~*fl~~*M~~~~B"J.~~~*m : WU i!i ~ & •. ~j (Bolles, 1972; Hull, 1943).
相關文件
volume suppressed mass: (TeV) 2 /M P ∼ 10 −4 eV → mm range can be experimentally tested for any number of extra dimensions - Light U(1) gauge bosons: no derivative couplings. =>
For pedagogical purposes, let us start consideration from a simple one-dimensional (1D) system, where electrons are confined to a chain parallel to the x axis. As it is well known
The observed small neutrino masses strongly suggest the presence of super heavy Majorana neutrinos N. Out-of-thermal equilibrium processes may be easily realized around the
incapable to extract any quantities from QCD, nor to tackle the most interesting physics, namely, the spontaneously chiral symmetry breaking and the color confinement..
(1) Determine a hypersurface on which matching condition is given.. (2) Determine a
• Formation of massive primordial stars as origin of objects in the early universe. • Supernova explosions might be visible to the most
• elearning pilot scheme (Four True Light Schools): WIFI construction, iPad procurement, elearning school visit and teacher training, English starts the elearning lesson.. 2012 •
(Another example of close harmony is the four-bar unaccompanied vocal introduction to “Paperback Writer”, a somewhat later Beatles song.) Overall, Lennon’s and McCartney’s
