Skip to main content
PMC home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2019 Oct 1.
Published in final edited form as: Psychon Bull Rev. 2018 Oct;25(5):1943–1951. doi: 10.3758/s13423-017-1402-9

A broader phenotype of persistence emerges from individual differences in response to extinction

Bruno Sauce 1, Christopher Wass 1, Michael Lewis 1, Louis D Matzel 1
PMCID: PMC5949244  NIHMSID: NIHMS920155  PMID: 29134544

Abstract

The typical practice of averaging group performance during extinction gives the impression that responding for an omitted reward declines gradually and homogeneously. However, previous studies of extinction in human infants have found that some individuals persist in responding while others abruptly cease responding. As predicted by theories on control, the infants who quickly resign typically display signs of sadness and despair when the expected reward is omitted. Using genetically diverse mice, here we observed a similar pattern of individual differences and associated phenotypes. After learning to approach a food reward, upon extinction, some animals rapidly abandoned approach to the goal box, while other animals persisted in entering and searching the goal box. The average performance of these two classes of mice suggested a smooth, gradual extinction curve, although a lack of homogeneity was revealed by a significant increase in variance around mean performance. Interestingly, the persistent mice were slower to “give up” when confined to an inescapable pool of water (a test asserted to be indicative of susceptibility to depression) and exhibited a more extensive pattern of search for omitted rewards. Thus, extinction reveals continuum in persistence, where low values might reflect a susceptibility to the negative effects of stress and might predispose the individuals to depression.

Keywords: extinction, mouse, persistence, individual differences, depression

1. Introduction

The omission of an expected reward is known as “extinction”, and this treatment leads to a decrease in goal directed behavior. Extinction curves (which typically illustrate average responses across groups of individuals) can obscure the variance between individuals, giving the impression that extinction is gradual and relatively homogeneous. In fact, extinction can occur at dramatically different rates across individuals (Andrews & Debus, 1978; Quinn, Brandon, & Copeland, 1996), and extensive work indicates that emotional responses to shifts in reward (including extinction) are related to dramatic personality differences (Corr, Pickering, & Gray, 1995; Gray, 1970).

Interestingly, persistence during extinction could be beneficial in many circumstances. A series of studies showed that during extinction, some human infants exhibit persistence accompanied by facial displays of anger (Alessandri, Sullivan, & Lewis, 1990; Crossman, Sullivan, Hitchcock, & Lewis, 2009; Sullivan, Lewis, & Alessandri, 1992) while others quickly stop responding and exhibit expressions of sadness (Lewis & Ramsay, 2005). These distinct individual differences were consistent from 4 to at least 24 months of age (Lewis, Sullivan, & Kim, 2015), and suggest the possibility of differences in a broader phenotype. Indeed, work by others suggests that children who express sadness in response to challenging situations often exhibit an absence of control, low self-worth, and a disposition for learned helplessness, all of which are associated with depression (Burhans & Dweck, 1995; Kistner, Ziegert, Castro, & Robertson, 2001).

Extinction has received tremendous attention in studies with non-human animals, and a number of studies have also assessed the emotional consequences of extinction, such as frustration from lack of reward, incentive downshifts, and extinction-induced aggression (Durlach, 1986; King, Scott, Graham, & Richardson, 2017; Matzel, 1984; Papini, 2014). However, little attention has been directed to how individual differences in persistence to extinction might be related to a broader phenotype, and there are no animal studies analogous to those with human infants described above. This is unfortunate, as individual differences in response to extinction may have important implications for such phenomenon as addiction relapse, resilience under stress, and susceptibility to depression.

Here we assessed individual differences in persistence in mice during extinction, and simultaneously monitored other traits that might be co-expressed with these differences such as susceptibility to depression, learning, and stress reactivity. The outbred mouse strain (CD-1) used here exhibit levels of behavioral and genetic variability comparable to wild mice (Aldinger, Sokoloff, Rosenberg, Palmer, & Millen, 2009), and thus are used extensively in studies of individual differences (Kolata, Light, & Matzel, 2008; Matzel et al., 2006, 2011; Wass et al., 2013).

2. Materials and Methods

2.1. Subjects

We used 26 outbred CD-1 male mice (Harlan Sprague Dawley Inc., Indianapolis, IN) that weighed 25–30 g and were approximately six weeks of age upon arrival in our laboratory. These mice express genetic variability comparable to that of wild-type mice, and have been observed to express a wide degree of behavioral variability. The mice were individually housed in clear shoebox cages inside a temperature-controlled colony room with a 12:12h light-dark cycle. To reduce the effect of individual differences in stress due to interactions with the experimenter, we handled all mice daily for 90 secs during the two weeks before the start of behavioral testing. (Handling consisted of holding a mouse on the palm of an experimenter’s hand, and systematically carrying it through the laboratory.) This handling is a routine procedure in our studies on individual differences in mice, and it leads to a noticeable reduction in mice’s defensive behaviors during tests. The mice were young adults (eight weeks of age; approximate equivalent of an 18-year-old human) at the start of testing, and had not participated in any other test before the current study.

2.2. Behavioral Tests

We measured the mice’s learning and extinction in a Long Alley. We also tested the same mice for anxiety/general arousal in an Open Field test, spatial learning in a Spatial Water Maze, and for their predisposition for depression using the Forced Swim test. The timeline for testing is illustrated in Figure 1. Since we were interested in individual differences, all mice were administered these tests in the same order. Except during testing in the Long Alley, all mice were fed ad libitum.

Figure 1.

Figure 1

Open in a new tab

The timeline of tests for a single run of mice: Open Field, Long Alley, Spatial Water Maze, and Forced Swim. Numbers indicate duration in days. Space between rectangles indicates resting time with no experimental manipulations.

2.2.1. Persistence during extinction

The Long Alley is a simple task that requires a mouse traverse a straight alley (from a start box) to a goal box to obtain a piece of food. The time to traverse the alley decreases over successive trials as a hungry mouse learns that food is located in the goal box. The apparatus we used was made of black Plexiglas and consisted of a long alley (112 cm long and 6.5 cm wide) with a small starting area (10 cm long) delimited by a remotely operated, vertical door. The alley ended in another remotely operated, vertical door, which led to a large, circular goal area (42 cm of diameter). On the side opposite its entry door, the goal area had a food cup on the Plexiglas floor formed by a 7-mm diameter, 5 mm deep depression.

For the acclimation to the Long Alley, we let each mouse freely explore the alley and goal box with doors opened and food absent for 12 min. For acquisition training, in each trial we placed a mouse in the start box for 20 sec. Then, all doors were opened and the mouse was allowed to run freely towards the goal area, which had the food cup on the floor baited with a piece of food (14 mg Noyes rodent grain pellet). When the mouse reached the goal area, the doors were closed and the mouse was confined there for 60 sec. We administered 10 days of acquisition training with six trials per day. From our previous experience, this level of training is sufficient to support a high level of efficacy and super-asymptotic performance (indicated by a reduction in running speed) in all mice. After completion of acquisition, we started the extinction phase of training that lasted for two days with six trials each day. The first two trials of the first day were exactly the same as during the acquisition phase (i.e., served as two additional acquisition training trials). For the next 10 trials, food was absent from the food cup. Ten days after the completion of extinction training, we tested the mice’s reacquisition of the learned response to obtain food. During this reacquisition phase, mice received trials identical to those during the initial acquisition phase. In all phases, we food deprived the mice by giving them only 90 min of access to food daily (delivered near the end of the mice’s light cycle), beginning on the day prior to training. This protocol leads to an average loss of 5% of the animals’ ad libitum body weight, which relative to many studies, would be a mild level of deprivation. During all phases, we measured the time (from the opening of the doors) for a mouse to reach the goal area. During the extinction phase, more persistent mice should traverse the alley faster than mice that more quickly abandon the previously learned running response.

2.2.2. Anxiety or general arousal

The Open Field is a commonly used test of the propensity for anxiety and/or stress reactivity, where mice are allowed to explore a novel, typically stress-inducing, open space (for a review on the topic, see Prut & Belzung, 2003). In a walled open field, rodents typically spend a majority of their time around the periphery of the field, and avoid entering the center (unwalled) areas of the field. Here, we used a 46 × 46 cm box with 20 cm high walls of white Plexiglas as the open field. The floor of the box was divided with tape creating a grid pattern to divide it in squares next to the outer walls of the field (i.e., “walled squares”), and squares in the interior of the field (i.e., “center squares”). The box was located in a brightly lit room in order to make the center squares even less appealing to explore, as mice are averse of open spaces and bright lights. We placed the mice in the center of the box and allowed them to explore it for 5 min, while recording for later scoring. As the measure of anxiety, we used the relative time (in percentage) spent in the walled squares. Thus, higher values indicate a preference for the walled periphery of the field, and suggest that the mouse found the center of the field anxiety or stress provoking.

2.2.3. Spatial learning

The Spatial Water Maze requires mice to locate a submerged platform in a pool of opaque water from which they are motivated to escape. With no explicit intramaze cues, mice’s performance in this maze is highly dependent on spatial cues located outside the pool (Morris, 1981). The path length to locate the platform typically decreases over successive trials, despite entering the pool from different locations at each trial. The apparatus consisted of a circular pool (diameter: 121 cm, depth: 42 cm) filled with water made opaque by nontoxic black paint. A black platform (10 cm of diameter) was hidden, submerged 3 cm under the water. The pool was inside a dark room, and it was surrounded by a black curtain with three bright, but differently shaped patterns of lights which served as visual cues.

For the acclimation to the pool, we confined each mouse to the pool’s platform (that was surrounded with a clear Plexiglas cylinder) for 4 min. For the acquisition phase, on each trial the mice were started from a different position by the wall (out of three possible positions). The platform was always in the same position. A mouse was said to have successfully located the platform when it remained on the platform for 5 sec. After locating the platform or swimming for 90 seconds, we left or placed mice on the platform for 5 secs, after which we removed them for a 10-min intertrial interval inside a warmed holding box. Each mouse completed three days of training with 7 trials per day, and was recorded for later scoring. We used the path length covered by each mouse inside the pool as the measure of spatial learning.

2.2.4. Forced Swim Test (the susceptibility to depression)

The Forced Swim test is commonly used to assess the propensity for depression in rodents. This test is based on principle that some individuals stop swimming (become immobile) when confined to an inescapable tank of water. The test has good predictive validity, as the majority of clinically used antidepressants decrease the duration of immobility (for a review on the topic, see Porsolt, Brossard, Hautbois, & Roux, 2001). We placed the mice individually in a vertical plastic cylinder containing water at 20°C (with 24°C air temperature) that was 30 cm diameter × 40 cm deep for 6 min. During this time, we measured the total duration of immobility, defined as the time a mouse floated passively in the water with no other movement beyond the occasional alternate movements of paws and tail necessary to keep head/nose above water. A higher duration of immobility correlates with higher tendency to depression (as the mouse “gives up” fighting for survival), and is indicative of higher sensitivity to treatments that induce depression-like symptoms (Petit-Demouliere, Chenu, & Bourin, 2005).

2.3. Serum levels of corticosterone during stress

One day after the end of behavioral tests, we measured the mice’s levels of corticosterone, an important stress hormone. First, in order to maximize individual differences in physiological stress response, we placed mice in a 50 mL Eppendorf tube for 10 min under a bright source of light (a procedure described as “restraint” stress). We then collected trunk blood by decapitation and allowed it to clot. The blood was centrifuged, and aliquots of serum were stored at − 80 °C, until analyzed. We used a commercially available enzyme immunoassay kit (Arbor Assays, K014-H1) following the manufacturers’ procedures to determine serum levels of corticosterone.

2.4. Statistical analyses

We defined the individual differences in persistence based on the two last trials of the extinction phase of the Long Alley (where the highest variance in persistence was observed; see below). We then performed two different types of analyses: 1) Linear regression analyses with individual differences in persistence as the predictor of anxiety/general arousal (Open Field), spatial learning (Spatial Water Maze), susceptibility to depression (Forced Swim), and stress response (serum levels of corticosterone after stress). These analyses included all mice in this study. 2) Two-tailed t-tests and ANOVAs to compare the groups at the two extreme sides of persistence. These groups consisted of the top 20% (n=5) most persistent mice and the bottom 20% (n=5). We used these thresholds based on the same percentages of top 20% and bottom 20% used in human studies described above, as setting this standard will make future comparison between species easier. We contrasted these two groups of mice on their performance in spatial learning (Spatial Water Maze), susceptibility to depression (Forced Swim), and in levels of stress response (serum levels of corticosterone after stress).

We also performed repeated measures ANOVAs, correlations, and power analyses, as described below. We used SPSS 21 for all analyses. The data in figures and text are expressed as mean and standard error of the mean. We considered a p value below 0.05 to indicate significance.

3. Results

3.1. Long Alley and persistence during extinction

For the Long Alley, we describe the performance during the initial phase of acquisition, during extinction, and again during re-acquisition. As seen in Figure 2A, mice, on average, exhibited a fast decline in the latency to reach the goal box. And by the end of training (Trial 63), there was very little variation in latencies across mice, which suggests that all mice learned the basic quickly and to a comparable degree. During extinction (Trials 64–74), the average latency to reach the goal increased substantially. Note also the large increase in variability observed during extinction, where a Mixed Model analysis showed that 38% of the variance in performance at the end of extinction was explained by individual differences alone. During reacquisition (starting 10 days after the completion of extinction training; Trials 75–80), latencies again decreased, ultimately reaching their pre-extinction values. Notably, mice reached asymptotic levels of performance with far less training after extinction relative to initial acquisition, i.e., they exhibited facilitated re-acquisition.

Figure 2.

Figure 2

Open in a new tab

A. Average time, in seconds, for all mice to reach the goal box during initial acquisition, extinction, and reacquisition. B. Average time, in seconds, for the 5 least persistent mice and the 5 most persistent mice identified during extinction. Brackets indicate standard error of the mean.

Figure 2B illustrates the performance of the five mice that were most or least persistent during extinction across all phases of training. Regarding the acquisition phase, we used Trials 1 to 24 (the approximate trial at which the mice started to reach asymptote) as a measure of their rate of learning. There were no differences in learning between the five least persistent mice and the five most persistent mice, F(1,7)=2.75, p=0.141; partial eta-squared=0.28. And, as expected, at the end of the acquisition phase (Trials 62–63), the two groups had similar latencies to reach the goal, t(8)=0.71, p>0.250; with an effect size of d=0.45. By the last two of the 10 extinction trials (Trials 73–74), the time to reach the goal was significantly different between the five least persistent mice and the five most persistent mice, t(8)=3.44, p=0.009; with an effect size of d=2.17. Note that while the most persistent mice exhibited no appreciable change in behavior during extinction, the least persistent mice exhibited rapid and complete extinction. In fact, by the end of extinction training, the least persistent mice were running slower than on the initial acquisition trials, t(4)=3.25, p=0.031. By the end of end of reacquisition (last two trials), these two populations were again performing similarly, exhibiting no differences in running speed and performing at a level comparable to that at the end of initial acquisition, t(7)=1.27, p=0.246; d=0.80.

During extinction training, we recorded the distance traveled in the circular goal box after a mouse encountered the empty food cup. Exploration of the goal box during an extinction trial might be regarded as an active search for the missing reward, and thus might serve as an independent measure of persistence. Although the extreme groups of the most and least persistent mice did not differ on this measure, distance traveled was significantly predicted by levels of persistence across all mice in the study, β=−0.45, t(25)=2.52, p=0.018.

3.2. Anxiety/General Arousal

Results for the Open Field are illustrated in Figure 3. Across all mice in the study, the relative time spent in the periphery (walled squares) was not significantly predicted by persistence, β=0.23, t(25)=1.20, p=0.241. Also, the five most persistent mice did not differ significantly from the five least persistent mice, t(8)=0.04, p>0.250; with a very small effect size, d=0.03.

Figure 3.

Figure 3

Open in a new tab

A. Individual values for all mice of time to reach the goal box (seconds) in the Long Alley over the time spent in the periphery (%) in the Open Field. The linear regression was not significant. B. Mean relative time spent in the periphery (%) of the Open Field for the 5 least persistent mice and the 5 most persistent mice. There was no significant difference between the two groups.

3.3. Spatial learning

Results for the Spatial Water Maze are illustrated in Figure 4. We used the mean distance traveled (path length) by each mouse from Trials 2 to 6 as a measure of the rate of spatial learning. These trials seem to be a good indicator of differences in rate of acquisition, and we have used them elsewhere for this purpose (Kolata, Wu, Light, Schachner, & Matzel, 2008; Sauce et al., 2015). As expected, we found that, across all mice, the rate of learning was not significantly predicted by persistence in the Long Alley, β=0.29, t(22)=1.421, p>0.250, R2=0.08. Furthermore, the rate of learning was not significantly different between the 5 most persistent and the 5 least persistent mice, t(8)=0.44, p>0.250; d=0.28. And at the end of the Spatial Water Maze, the 5 least persistent mice seem to travel the same distance relative to the 5 most persistent. Together with the results from the Long Alley, this data provides further evidence that the differences in persistence during extinction are not a reflection of differences in learning.

Figure 4.

Figure 4

Open in a new tab

A. Average distance traveled (path length, in cm) in the Water Maze for all mice. B. The 5 least persistent mice and the 5 most persistent mice during extinction in the Straight Alley were compared across 21 trials in the water maze. There were no significant differences in rate of learning or asymptotic performance between the two groups. Brackets indicate standard error of the mean.

3.4. Forced Swim Test (susceptibility to depression)

Results for the Forced Swim are illustrated in Figure 5. Across all mice in the study, the duration of immobility was significantly predicted by persistence in the Long Alley, β=0.46, t(22)=2.45, p=0.023, such that mice that were more persistent during extinction also exhibited more persistent swimming, i.e., they were less likely to become passive That regression model explained 21% of the total variance in immobility, R2=0.21, F(1, 22)=6.00, p=0.023. And in the contrast of only the two extreme groups, the five most persistent mice during extinction exhibited significantly more mobility than the least persistent mice, t(8)=2.58, p=0.033; with a large effect size, d=1.82 (see Figure 4).

Figure 5.

Figure 5

Open in a new tab

A. Individual values for all mice of time to reach the goal box (seconds) in the Long Alley over the duration of immobility (seconds) during the Forced Swim. The linear regression was significant. B. Mean duration of immobility (seconds) during the Forced Swim for the five least persistent mice (during extinction in the Straight Alley) and the five most persistent mice. * = p < .05. Brackets indicate standard error of the mean.

3.5. Serum corticosterone levels during stress

Across all mice in the study, the corticosterone response to a stressful situation (forced restraint) was not significantly predicted by persistence during extinction in the Long Alley, β=−0.02, t(22)=0.07, p>0.250. And in the group comparison of the extreme groups, the 5 most persistent mice did not differ significantly from the 5 least persistent mice, t(8)=0.40, p>0.250; with a small effect size, d=0.28 (see Figure 6B).

Figure 6.

Figure 6

Open in a new tab

A. Individual values for all mice of time to reach the goal box (seconds) in the Long Alley over the serum corticosterone concentration after an acute stress. The linear regression was not significant. B. Mean serum corticosterone concentration after an acute stress for the five least persistent mice and the five most persistent mice. There were no significant differences between the two groups.

4. Discussion

Extinction “curves” based on averages support the impression that individuals exhibit a slow and gradual decline in goal directed behavior after the removal of reward. Here, genetically heterogeneous mice differed dramatically in their persistence in response during extinction. Some quickly suppressed responding, while others exhibited very high persistence. These mice also presented other patterns of behavior that covaried with persistence, and the extremes of the continuum were quite different in key phenotypes.

Differences in response to extinction did not seem to be a product of differences in the rate at which the mice learn. Mice that were more persistent during extinction were not better at new learning of spatial navigation or the Long Alley task. Hence, individual differences in extinction might reflect differences in emotionality as found in human infants (Lewis & Ramsay, 2005). Similarly, there is a possibility that factors such as general arousal or temperament might play a role in the rate at which an individual extinguishes goal-directed responding. Our results with the open field here, however, suggest that this may not be the case. Mice’s behavior in the open field can usually reveal difference in emotionality or anxiety, but our persistent mice did not differ from the non-persistent ones in their patterns of behavior in the open field.

A positive correlation was observed between mice’ level of persistence during extinction and the amount of search behavior engaged in after encountering the empty food cup. Moreover, mice that were more persistent during extinction continued swimming in a pool of inescapable water after the low persistence mice had “given up” (i.e., resorted to passive floating). The forced swim test is considered a reliable predictor of the susceptibility to depression in mice, and is a common screening method for the efficacy of putative antidepressant drugs (Petit-Demouliere, Chenu, & Bourin, 2005). Thus, the phenotype that emerges in response to extinction may be indicative of susceptibility to some forms of depression. Our results with mice are remarkably similar to the observation that human infants who rapidly suppress responding during extinction display sadness and despair when the expected reward is withheld (Kistner et al., 2001; but see Cloninger, Zohar, Hirschmann, & Dahan, 2012), and persistence training may inoculate individuals against depression (Nation & Massad, 1978). It is important to note that our study only used male animals, and so our conclusions on persistence and its broader phenotype are sex specific. This may be particularly relevant regarding predisposition to depression, the etiology of which is probably different in males and females.

Here, persistence was unrelated to glucocorticoid levels in response to environmental stress. This contrasts with results obtained in children, where those who rapidly abandon responding during extinction exhibit sadness and higher levels of cortisol after a stressor (Lewis & Ramsay, 2005). Relatedly, the high and low persistence mice tested here did not differ in their patterns of behavior in an open field, a test which can be sensitive to differences in anxiety or stress reactivity. Thus, it does not appear that persistence need necessarily be related to differences in stress reactivity, and instead, may emerge independently.

An alternative explanation to the one we put forward here is that the positive correlation between persistence and forced swim is not indicative of persistence per se, but of an underlying physiological trait such as vigor or a motivational state. However, we believe this is unlikely to be the case, since behavior in the Open Field (which is sensitive to general arousal) was unrelated to persistence. Moreover, a motivational state is likely to impact acquisition and/or asymptotic performance during learning, and in the present experiment, no relationship between learning and persistence was observed. In total, the present results suggest that differences in persistence emerge independently of motivation or arousal.

The present results suggest that the continuum in persistence revealed in response to extinction reflect broader phenotypic differences. Furthermore, a persistent phenotype might be protective and lead to more resilience. Some humans exposed to stressful events never exhibit psychopathology such as posttraumatic stress disorder or depression (Yehuda, 2004). These “resilient” individuals display traits such as cognitive flexibility and optimism (Charney, 2004). This can be further elucidated in mice through environmental manipulations. For instance, exposure to novel environments promotes increases in exploratory behaviors (Franks, Champagne, & Higgins, 2013; Light, Kolata, Hale, Grossman, & Matzel, 2008) and might transform mice that “give up” into persistent ones. Such studies could aid our understanding of the interplay of extinction, persistence, control, and depression in humans. Furthermore, persistence as revealed by extinction might provide insights into individual sensitivities to antidepressant drugs, which are often effective in only a small percentage of recipients (Kirsch, 2008).

Acknowledgments

This work was supported by a grant from the National Institute of Mental Health (AG022698) and the Office of Naval Research (MH108706) to L.D.M.

References

  1. Aldinger Ka, Sokoloff G, Rosenberg DM, Palmer Aa, Millen KJ. Genetic variation and population substructure in outbred CD-1 mice: implications for genome-wide association studies. PloS One. 2009;4(3):e4729. doi: 10.1371/journal.pone.0004729. [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Alessandri SM, Sullivan MW, Lewis M. Violation of expectancy and frustration in early infancy. Developmental Psychology. 1990;26(5):738–744. doi: 10.1037/0012-1649.26.5.738. [DOI] [Google Scholar]
  3. Andrews GR, Debus RL. Persistence and the causal perception of failure: Modifying cognitive attributions. Journal of Educational Psychology. 1978;70(2):154–166. doi: 10.1037/0022-0663.70.2.154. [DOI] [Google Scholar]
  4. Burhans KK, Dweck CS. Helplessness in Early Childhood: The Role of Contingent Worth. Child Development. 1995;66(6):1719. doi: 10.2307/1131906. [DOI] [PubMed] [Google Scholar]
  5. Charney DS. Psychobiological Mechanisms of Resilience and Vulnerability. FOCUS. 2004;2(3):368–391. doi: 10.1176/foc.2.3.368. [DOI] [PubMed] [Google Scholar]
  6. Cloninger CR, Zohar AH, Hirschmann S, Dahan D. The psychological costs and benefits of being highly persistent: Personality profiles distinguish mood disorders from anxiety disorders. Journal of Affective Disorders. 2012;136(3):758–766. doi: 10.1016/j.jad.2011.09.046. [DOI] [PubMed] [Google Scholar]
  7. Corr PJ, Pickering AD, Gray JA. Personality and reinforcement in associative and instrumental learning. Personality and Individual Differences. 1995;19(1):47–71. doi: 10.1016/0191-8869(95)00013-V. [DOI] [Google Scholar]
  8. Crossman AM, Sullivan MW, Hitchcock DM, Lewis M. When frustration is repeated: Behavioral and emotion responses during extinction over time. Emotion. 2009;9(1):92–100. doi: 10.1037/a0014614. [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Durlach PJ. Explicitly unpaired procedure as a response elimination technique in autoshaping. Journal of Experimental Psychology: Animal Behavior Processes. 1986;12(2):172–185. doi: 10.1037/0097-7403.12.2.172. [DOI] [Google Scholar]
  10. Franks B, Champagne FA, Higgins ET. How Enrichment Affects Exploration Trade-Offs in Rats: Implications for Welfare and Well-Being. PLoS ONE. 2013;8(12):e83578. doi: 10.1371/journal.pone.0083578. [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Gray JA. The psychophysiological basis of introversion-extraversion. Behaviour Research and Therapy. 1970;8(3):249–266. doi: 10.1016/0005-7967(70)90069-0. [DOI] [PubMed] [Google Scholar]
  12. King G, Scott E, Graham BM, Richardson R. Individual differences in fear extinction and anxiety-like behavior. Learning & Memory. 2017;24(5):182–190. doi: 10.1101/lm.045021.117. [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Kirsch I. Antidepressant drugs “work”, but they are not clinically effective. British Journal of Hospital Medicine. 2008;69(6):359–359. http://doi.org/10.12968/hmed.2008.69.6.29632. [PubMed] [Google Scholar]
  14. Kistner JA, Ziegert DI, Castro R, Robertson B. Helplessness in Early Childhood: Prediction of Symptoms Associated With Depression and Negative Self-Worth. Merrill-Palmer Quarterly. 2001;47(3):336–354. doi: 10.1353/mpq.2001.0016. [DOI] [Google Scholar]
  15. Kolata S, Light K, Matzel LD. Domain-Specific and Domain-General Learning Factors are Expressed in Genetically Heterogeneous CD-1 mice. Intelligence. 2008;36(6):619–629. doi: 10.1016/j.intell.2007.12.001. [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Kolata S, Wu J, Light K, Schachner M, Matzel LD. Impaired working memory duration but normal learning abilities found in mice that are conditionally deficient in the close homolog of L1. The Journal of Neuroscience : The Official Journal of the Society for Neuroscience. 2008;28(50):13505–13510. doi: 10.1523/JNEUROSCI.2127-08.2008. [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Lewis M, Ramsay D. Infant Emotional and Cortisol Responses to Goal Blockage. Child Development. 2005;76(2):518–530. doi: 10.1111/j.1467-8624.2005.00860.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Lewis M, Sullivan MW, Kim HMS. Infant approach and withdrawal in response to a goal blockage: Its antecedent causes and its effect on toddler persistence. Developmental Psychology. 2015;51(11):1553–63. doi: 10.1037/dev0000043. [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Light K, Kolata S, Hale G, Grossman H, Matzel LD. Up-regulation of exploratory tendencies does not enhance general learning abilities in juvenile or young-adult outbred mice. Neurobiology of Learning and Memory. 2008;90(2):317–29. doi: 10.1016/j.nlm.2008.04.007. [DOI] [PubMed] [Google Scholar]
  20. Matzel LD. Aggression in rats as a function of target location in a double alley. The American Journal of Psychology. 1984;97(4):519–526. [PubMed] [Google Scholar]
  21. Matzel LD, Light K, Wass C, Colas-Zelin D, Denman-Brice A, Waddel AC, Kolata S. Longitudinal attentional engagement rescues mice from age-related cognitive declines and cognitive inflexibility. Learning & Memory (Cold Spring Harbor NY) 2011;18(5):345–56. doi: 10.1101/lm.2034711. [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Matzel LD, Townsend Da, Grossman H, Han YR, Hale G, Zappulla M, Kolata S. Exploration in outbred mice covaries with general learning abilities irrespective of stress reactivity, emotionality, and physical attributes. Neurobiology of Learning and Memory. 2006;86(2):228–40. doi: 10.1016/j.nlm.2006.03.004. [DOI] [PubMed] [Google Scholar]
  23. Morris RGM. Spatial localization does not require the presence of local cues. Learning and Motivation. 1981;12:239–260. doi: 10.1016/0023-9690(81)90020-5. [DOI] [Google Scholar]
  24. Nation JR, Massad P. Persistence training: A partial reinforcement procedure for reversing learned helplessness and depression. Journal of Experimental Psychology: General. 1978;107(4):436–451. doi: 10.1037/0096-3445.107.4.436. [DOI] [PubMed] [Google Scholar]
  25. Papini MR. Diversity of Adjustments to Reward Downshifts in Vertebrates. International Journal of Comparative Psychology. 2014;27(3) [Google Scholar]
  26. Petit-Demouliere B, Chenu F, Bourin M. Forced swimming test in mice: a review of antidepressant activity. Psychopharmacology. 2005;177(3):245–255. doi: 10.1007/s00213-004-2048-7. [DOI] [PubMed] [Google Scholar]
  27. Porsolt RD, Brossard G, Hautbois C, Roux S. Current Protocols in Neuroscience. Hoboken, NJ, USA: John Wiley & Sons, Inc; 2001. Rodent Models of Depression: Forced Swimming and Tail Suspension Behavioral Despair Tests in Rats and Mice. Vol Chapter 8, p Unit 8.10A. [DOI] [PubMed] [Google Scholar]
  28. Prut L, Belzung C. The open field as a paradigm to measure the effects of drugs on anxiety-like behaviors: a review. European Journal of Pharmacology. 2003;463(1–3):3–33. doi: 10.1016/S0014-2999(03)01272-X. [DOI] [PubMed] [Google Scholar]
  29. Quinn EP, Brandon TH, Copeland AL. Is task persistence related to smoking and substance abuse? The application of learned industriousness theory to addictive behaviors. Experimental and Clinical Psychopharmacology. 1996;4(2):186–190. doi: 10.1037/1064-1297.4.2.186. [DOI] [Google Scholar]
  30. Sauce B, Wass C, Netrakanti M, Saylor J, Schachner M, Matzel LD. Heterozygous L1-deficient mice express an autism-like phenotype. Behavioural Brain Research. 2015;292:432–442. doi: 10.1016/j.bbr.2015.05.040. [DOI] [PubMed] [Google Scholar]
  31. Sullivan MW, Lewis M, Alessandri SM. Cross-age stability in emotional expressions during learning and extinction. Developmental Psychology. 1992;28(1):58–63. doi: 10.1037/0012-1649.28.1.58. [DOI] [Google Scholar]
  32. Wass C, Pizzo A, Sauce B, Kawasumi Y, Sturzoiu T, Ree F, Matzel LD. Dopamine D1 sensitivity in the prefrontal cortex predicts general cognitive abilities and is modulated by working memory training. Learning & Memory (Cold Spring Harbor NY) 2013;20(11):617–627. doi: 10.1101/lm.031971.113. [DOI] [PMC free article] [PubMed] [Google Scholar]
  33. Yehuda R. Risk and resilience in posttraumatic stress disorder. The Journal of Clinical Psychiatry. 2004;65(Suppl 1 (1)):29–36. Retrieved from http://www.ncbi.nlm.nih.gov/pubmed/14728094. [PubMed] [Google Scholar]

RESOURCES