Cannabis use is associated with sexually dimorphic changes in executive control of visuospatial decision-making

Abstract

When the outcome of a choice is less favorable than expected, humans and animals typically shift to an alternate choice option on subsequent trials. Several lines of evidence indicate that this "lose-shift" responding is an innate sensorimotor response strategy that is normally suppressed by executive function. Therefore, the lose-shift response provides a covert gauge of cognitive control over choice mechanisms. We report here that the spatial position, rather than visual features, of choice targets drives the lose-shift effect. Furthermore, the ability to inhibit lose-shift responding to gain reward is different among male and female habitual cannabis users. Increased self-reported cannabis use was concordant with suppressed response flexibility and an increased tendency to lose-shift in women, which reduced performance in a choice task in which random responding is the optimal strategy. On the other hand, increased cannabis use in men was concordant with reduced reliance on spatial cues during decision-making, and had no impact on the number of correct responses. These data (63,600 trials from 106 participants) provide strong evidence that spatial-motor processing is an important component of economic decision-making, and that its governance by executive systems is different in men and women who use cannabis frequently.

Keywords: addiction; cannabis; choice; executive control; habit; lose-shift; sex differences; spatial processing.

Figure 1

Behavioral task. (A) Timeline of trials in the matching pennies game. (B) Reconfiguration of targets, which could undergo local (swap and displace) and global changes in position.

Figure 2

Effect of recreational drug use on measures of task performance in males and females: proportion of wins (A); response entropy (B); and decision times (C). Plots show the conventional descriptive statistics: mean (diamond), median (horizontal line in the box), 25th/75th percentiles (box edges), and outliers (dots). ^*p < 0.05, ^**p < 0.01.

Figure 3

Effects of target reconfiguration and recreational drug use on reinforcement-driven behavior. (A,B) Effect of local and global changes in choice position on lose-shift and win-stay tendencies for all participants. SEM in error bars. (C,D) Box plots of the difference in lose-shift and win stay when target positions are swapped as compared to no change. The effect of swapping targets on lose-shift is higher in women who use cannabis, but lower in men who use cannabis, than their sex-matched controls. (E) Correlation between total drug use (ASSIST) and swap effect on lose-shift. *p < 0.05.

Figure 4

Effect of local and global changes in choice position on lose-shift tendencies in male and female cannabis users, relative to controls. SEM in error bars.

Figure 5

(A) Performance of the Q-learning (Q), Q-learning with forgetting (FQ), and Q-learning with differential forgetting (DFQ) models. (B) Relationship between κ₁ and win-stay behavior with curves fit to individual subjects (blue lines), the population average (black line), and averages of binned raw data (points). (C,D) Relationship between κ₂ & α and lose-shift behavior. α was normalized via the logit transform prior to model fitting.

Figure 6

Effect of wins, local, and global changes in choice position on Q-learning parameters α (A) (learning rate), κ₁ (B) (reward strength), and κ₂ (C) (punishment strength). The influence of wins and cue rearrangement during the previous 32 trials is estimated by Volterra decomposition, which provides a weight (loading) for each trial lag. SEM in shaded area.

Figure 7

Cannabis × Sex interactions on mean reinforcement learning parameter values estimated by Volterra decomposition. Learning rate α (A) and effect of wins κ₁ (B) to wins, local displacement, swaps, and global changes. SEM in error bars. *p < 0.05, **p < 0.01.