BTBR Mice Exhibit Deficits In Probabilistic Reversal Learning

Background: The BTBR T⁺ tf/J (BTBR) mouse exhibits repetitive motor behaviors  and restricted interests similar to that observed in autism spectrum disorder (ASD). Less clear is whether the BTBR mouse displays cognitive flexibility deficits also observed in ASD.      

Objectives:  The present experiment investigated whether BTBR mice compared to C57BL/6J mice exhibit impairments in acquisition and/or reversal learning of a spatial discrimination using a completely accurate feedback procedure vs. a probabilistic learning procedure.  In addition, both mouse strains were tested on different measures of motor stereotypy.   

Methods:  

In Experiment 1, BTBR mice and C57BL6J mice were tested  on  acquisition and reversal learning of a two-choice spatial discrimination using 100% accurate feedback.  In Experiment2,  the two mouse strains were tested on acquisition and reversal learning of a spatial discrimination using a 80/20 probabilistic learning procedure. In Experiment 2, mice were also tested for  repetitive grooming and marble burying, as well as spontaneous alternation as a measure of short-term memory. In the spatial discrimination,  mice were trained to obtain a cereal reinforcement  from one of two food wells  placed in distinct locations within a rectangular-shaped maze. Testing occurred across two consecutive days.  The learning criterion for the acquisition and reversal learning phase were each 6 consecutive trials.  In Experiment 1, 100% accurate feedback was provided.  In Experiment 2, the “correct” choice was reinforced on 80% of trials and the “incorrect” choice was reinforced on 20% of trials.  To measure grooming mice were placed in a clear acrylic chamber for 10 min and the duration of grooming was measured. In the marble burying test, mice were placed in a plastic chamber with wood chip bedding and the presence of 20 marbles for 30 min undisturbed; buried marbles were recorded.  In the spontaneous alternation test, mice explored a three- arm maze undisturbed for 20 min, while entry patterns were recorded. 

Results:  BTBR and  C57BL6J mice performed comparably on acquisition and reversal learning of a spatial discrimination with 100% accurate feedback.  BTBR mice performed similarly as C57BL6J mice on spatial acquisition, but were impaired in spatial reversal learning using a probabilistic learning procedure.  The deficit was a result from a significant increase in regressive errors in which BTBR mice were impaired in maintaining the new choice pattern after being initially selected.   As previously observed, BTBR mice spent significantly more time grooming and buried significantly more marbles compared to C57BL6J mice. In contrast,  BTBR and C57BL6J mice exhibited similar spontaneous alternation scores.   not differ in the trials needed to reach criterion during place acquisition. Instead, analyses indicate that BTBR mice required significantly more trials to inhibit responding for the previously “correct” arm and switching to a new choice pattern.

Conclusions:  The present findings indicate that BTBR mice not only exhibited increased repetitive motor behaviors, but also exhibit cognitive flexibility deficits as measured by probabilistic reversal learning.  Thus, BTBR mice can serve as a useful model to study the mechanisms and potential treatments for cognitive flexibility deficits in ASD.