Heading perception is a multi-sensory process that involves both visual (e.g. optic flow) and vestibular (e.g. inertial motion) signals. We previously reported that activity in extrastriate visual cortex (area MSTd) is correlated with monkeys’ perceptual judgments during performance of a heading discrimination task based on visual or vestibular cues. To test for a causal link between MSTd activity and heading discrimination, we used microstimulation and reversible inactivation techniques to artificially manipulate MSTd activity and observe the subsequent changes in the monkey’s behavioral performance.

Methods: Heading stimuli were defined by optic flow, inertia motion or congruent combination of the two cues. Heading direction was varied in fine steps around straight forward in the horizontal plane. Monkeys were required to report whether their perceived heading was to the left or right of straight ahead. For microstimulation, during one-half of the trials, a mild electrical current (peak amplitude: 20 uA) was applied through a microelectrode placed in area MSTd. For inactivation, a small amount of muscimol (10mg/ml, 1~2ul) was injected through a canula placed in area MSTd. Chemical injections were applied bilaterally.

Results: 1, Microstimulation: In the visual condition, microstimulation of MSTd produced a statistically significant behavioral bias at 81/150 (54%) stimulation sites. Among the significant effects, the majority (86%) were in the direction predicted from the heading tuning of multi-unit activity at the stimulation site. At the population level, both monkeys showed a highly significant effect (p<<0.001). In the vestibular condition, microstimulation elicited a significant behavioral bias for only 16/150 (10.7%) sites. Among the significant effects, about half were consistent with the heading preference of multiunit activity. At the population level, neither monkey showed a significant stimulation effect (p>0.1). An examination of heading tuning for multiunit activity revealed substantially weaker vestibular selectivity than visual selectivity, suggesting that weak clustering of inertial motion signals in area MSTd limits the efficacy of microstimulation in the vestibular condition. In the combined condition, the effect was intermediate between the two single cue conditions. 2, Inactivation: In the visual condition, inactivating MSTd significantly deteriorated the monkeys’ heading discrimination performance (p<<0.001, n=13). The average psychophysical threshold increased about 3-fold compared to that in the control and recovery sessions. In the vestibular condition, the monkeys’ performance was also significantly affected (p<0.001, n=13) after inactivation. However, the average threshold increased by only 30%, a much weaker effect than in the visual condition. In the combined condition, the inactivation effect was significant (p<0.001, n=13) and the average threshold increased by about 30%.

Conclusion: These results establish a causal link between MSTd activity and monkeys’ heading perception based on visual and vestibular signals. However, based on the behavioral effects of microstimulation and inactivation, MSTd appears to play a more dominant role in heading judgments from visual signals alone than vestibular signals alone. The weak vestibular effects suggest that other areas may play significant roles in judging self-motion from vestibular cues.