Motor effort alters changes of mind in sensorimotor decision making

Abstract

After committing to an action, a decision-maker can change their mind to revise the action. Such changes of mind can even occur when the stream of information that led to the action is curtailed at movement onset. This is explained by the time delays in sensory processing and motor planning which lead to a component at the end of the sensory stream that can only be processed after initiation. Such post-initiation processing can explain the pattern of changes of mind by asserting an accumulation of additional evidence to a criterion level, termed change-of-mind bound. Here we test the hypothesis that physical effort associated with the movement required to change one's mind affects the level of the change-of-mind bound and the time for post-initiation deliberation. We varied the effort required to change from one choice target to another in a reaching movement by varying the geometry of the choice targets or by applying a force field between the targets. We show that there is a reduction in the frequency of change of mind when the separation of the choice targets would require a larger excursion of the hand from the initial to the opposite choice. The reduction is best explained by an increase in the evidence required for changes of mind and a reduced time period of integration after the initial decision. Thus the criteria to revise an initial choice is sensitive to energetic costs.

Figure 1. Experimental set-up.

A: Schematic of the visual display (rectangle). A trial starts when the subject's hand is in the home position. After a random delay, the random dot kinematogram become visible and the subject views the moving dot stimulus for as long as they need (up to 2 s). Subjects indicate the direction of dot motion by moving to the leftward or rightward choice target. As soon as the subject moves from the home position, the motion stimulus vanished. The trial ended when the subject reached one of the two choice targets. B: Subjects held the handle of a robotic interface and moved to either a leftward or a rightward circular target depending on the perceived motion direction of a central random-dot display. A mirror system prevented subjects from seeing their arm. C: The time course of events that make up a trial.

Figure 2. Schematic of a model that explains the pattern of changes of mind .

Information flow diagram showing visual stimulus and events leading to a decision and a possible change of mind. The example illustrates a leftward motion stimulus that gives rise to an initial incorrect rightward choice. The visual stimulus gives rise to a decision variable that reflects accumulated evidence (black trace) that is the integral of noisy evidence. This governs the initial choice and decision time. The initial decision is complete when a ‘Right’ or ‘Left’ bound is crossed. There is a sensory delay from motion onset to the beginning of the accumulation and a motor delay from the initial decision to movement initiation. These delays together are termed non-decision time (t_nd). After the initial decision, further accumulation takes place on the evidence still in the processing pipeline; if the accumulated evidence reaches the opposite change-of-mind bound (B_com) within a temporal deadline (t_pip) then the decision is reversed (red). Failure to reach the change-of-mind bound (e.g. black trace) leads to no change of mind. Note that due to the time delays, only the yellow part of the visual stimulus can influence the initial choice and the blue portion can only be processed after the initial choice.

Figure 3

A: Sample hand trajectories from Subject 2 for the four conditions (3 different angular target separations and force field FF). Most trajectories extend directly from the home position (bottom circle) to one of the choice targets. B: In a fraction of trials, the trajectories change course during the movement, indicating a change of mind. Panel B shows all the change of mind trials for this subject and in panel A and equal number of non change-of-mind trials have been randomly selected. Note the visual/haptic wedge just above the home position which subjects were not allowed to contact.

Figure 4. Psychometrics of choice for all subjects for the 30° condition.

A: Reaction time as a function of motion coherence (specified as the proportion of dots moving in the same direction). Open circles are mean reaction times and solid lines are fits of the data to the drift-diffusion model. B: Proportion of correct trials as a function of motion coherence. The solid lines are the fits of the data. C: Proportion of trials with changes of mind as a function of motion coherence. Solid circles are for all changes of mind and open circles are the proportion of trials on which a change of mind improved performance (that is the difference in the number of changes of mind that correct an error and the number that induce an error). The solid and dotted lines are the fits of the extension of the drift-diffusion model for changes of mind to the data. Error bars show ±one standard error and are derived from the binomial distribution for the proportions.

Figure 5. Changes of minds.

A. Proportion of trials with a change of mind as a function of the condition for each subject. B: Proportion improvement on change of mind trials - that is the ratio of change of mind trials that correct an error to the total number of change of mind trials. Values exceeding one half (red bar) are consistent with an overall tendency to correct an initial error. Error bars are s.e.

Figure 6. Initial choice function.

Graphs show the probability of a rightward choice as a function of motion strength (sign of coherence reflects dot motion direction: left or right) for each subject and condition. Curves are logistic fits. For comparison the red curve for each subject is the logistic fit to their 15° condition. Error bars are s.e.

Figure 7. Pattern of changes of mind for Subject 2.

Probability of a change of mind as a function of (unsigned) motion coherence for the four conditions. Green circles are all change of mind trials, red are initial errors that were corrected and blue are trials that spoiled an initially correct choice. Solid curves are model fits to the data for the change of mind model. Error bars are s.e.

Figure 8. Parameter fits to the change of mind model for each subject.

The change of mind model fits the change of mind bound B_com and the time allowed for post-initiation processing t_pip. For each condition, the maximum likelihood estimates (small solid circles) are shown, as well as the 95% confidence region for the parameter estimate (shaded region). The proportion of change-of-mind isocontour lines are shown for the proportion of change of mind trials in the 15° target separation condition (solid line) as well as for proportion of change of mind trials that are 33, 67 and 133% that of the 15° separation condition.