Size and reversal learning in the beagle dog as a measure of executive function and inhibitory control in aging

Abstract

Several studies converge on the idea that executive processes age earlier than other cognitive processes. As part of a larger effort to investigate age-related changes in executive processes in the dog, inhibitory control was measured in young, middle-aged, old, and senior dogs using size discrimination learning and reversal procedures. Compared to young and middle-aged dogs, old and senior dogs were impaired on both the initial learning of the size task and the reversal of original reward contingencies. Impaired performance in the two aged groups was characterized as a delay in learning the correct stimulus-reward contingencies and, among the senior dogs in particular, an increase in perseverative responding. These separate patterns of reversal impairments in the old and senior dogs may reflect different rates of aging in subregions of the frontal cortex.

Figure 1

Box plots of total errors (A) and trials (B) to criterion for young, middle-aged, old, and senior dogs on initial size discrimination learning. Median errors and trials to criterion for each group are indicated by a single line inside the box plot. Longer box plots and whiskers indicate greater variability and skewness respectively, within a group. Individual data points indicate outliers within a group.

Figure 2

Backward learning curves for the acquisition of the size discrimination task by young (A), middle-aged (B), old (C), and senior (D) dogs. Data points represent mean percent correct scores (± standard error of the mean) and are plotted backwards from the test session in which criterion is reached (i.e., 0).

Figure 3

Rate of learning comparisons for total errors (A) and trials (B) to criterion on size and size reversal discrimination tasks. Young, old, and senior dogs took longer to learn the reversal task compared to the original size-learning condition.

Figure 4

Box plots of total errors (A) and trials (B) to criterion for young, middle-aged, old, and senior dogs on size-reversal discrimination learning. Median errors and trials to criterion for each group are indicated by a single line inside the box plot. Longer box plots and whiskers indicate greater variability and skewness respectively, within a group. Individual data points indicate outliers within a group.

Figure 5

Median number of sessions spent at stage I, II, and III during reversal learning by young, middle-aged, old, and senior beagle dogs.