Effort cost of harvest affects decisions and movement vigor of marmosets during foraging

Abstract

We prefer to decline effortful rewards, but if the circumstances require it, we will move only slowly to harvest them. Why should economic variables such as reward and effort affect movement vigor? In theory, our decisions and movements both contribute to a measure of fitness in which the objective is to maximize rewards minus efforts, divided by time. To test this idea, we engaged marmosets in a foraging task in which on each trial they decided whether to work by making saccades to visual targets, thus accumulating food, or to harvest by licking. We varied the effort cost of harvest by moving the food tube with respect to the mouth. Theory predicted that the subjects would respond to the increased effort costs by working longer, stockpiling food before commencing harvest, but reduce their movement vigor to conserve energy. Indeed, in response to the increased effort costs of harvest, marmosets increased their work duration but reduced their movement vigor. These changes in decisions and movements coincided with changes in pupil size. As the effort cost of harvest declined, work duration decreased, the pupils dilated, and lick and tongue vigor increased. Thus, when acquisition of reward became effortful, there was a global change in the state of the brain: the pupils constricted, the decisions exhibited delayed gratification, and the movements displayed reduced vigor.

Control of tongue movements by the Purkinje cells of the cerebellum

“…This food was presented via either the left or the right tube for 50-300 consecutive trials, and then switched tubes. Because the amounts were small (0.015-0.02 mL), and the tubes were located at ±90 o with respect to the mouth, harvesting was effortful (22), requiring skillful movements toward a target that was just large enough to accommodate the tongue (4.4 mm diameter tube). As a result, the subjects naturally chose to work for a few consecutive saccade trials (n=6.1±0.02 successful trials per work period), allowing the food to accumulate, then stopped making saccades to targets and initiated their harvest via a bout of licking (n=22.03 ± 0.04 licks per harvest period, Fig.…”

“…6B). To test if this modulation varied with tongue kinematics, we quantified lick vigor, defined as the peak speed of a given protraction with respect to the speed expected for a lick of the same amplitude (22,26) (Supplementary Figure 5). High vigor licks exhibited greater SS rates at peak protraction speed (Fig.…”

The olivary input to the cerebellum dissociates sensory events from movement plans

“…As the motivational state of the subject changed, so did saccade peak velocity ( 35 ). We measured peak velocity as a function of saccade amplitude in each subject and then normalized the data to compute vigor of each saccade ( 36 , 37 ).…”

A collicular map for touch-guided tongue control

“…Altogether, these data show that the latSC contains a mechanosensorimotor map for touch-guided tongue control that has many similar mechanistic features observed in visually-guided orienting 24,25 . With recent advances examining tongue kinematics in marmosets, it will be possible to test if primate latSC also implements touch-guided control 54 .…”