Inactivation in LIP disrupts behavior in a control task

Although inactivation in LIP had no measurable effect on direction discrimination, it did exert measurable effects on a “free-choice” control task, which was performed on every inactivation session (Fig. 3.6). Inactivation of LIP biased choices away from the contralateral hemifield (8.88% reduction from baseline on average, t(33) = 3.4, p = 0.001, paired t test), (Fig. 3.7), consistent with previous reports in monkeys (Wardak et al. 2004, Balan & Gottlieb 2009, Wilke et al. 2012), rodents (Erlich et al. 2015), and parietal lesions in humans (Kerkhoff 2001). Thus, our standard electrophysiological confirmation of LIP inactivation was complemented by a behavioral conse- quence in this free-choice control task. In addition to exerting a spatial bias, LIP inactivation caused an increase in endpoint error of saccades made to the hemifield contralateral to the inactivation (0.36◦ on average, t(33) = 4.4, p < 0.0001, Fig. 3.7).

Targets

flash (0.2s) Memory guidedfree choice Fixation

Targets Reward Eye x, y

a

b

Figure 3.6: Design of the free-choice control task. A, The “free-choice” task. Following a 200ms long presentation of two targets at random locations in space, monkeys were required to hold fixation for another 600-3,000ms, and then to move their eyes to the remembered location of either target. B, Task timing. Events in the task were presented in sequence and were jittered in time (gray arrows).

a

b

Difference in saccade error (°)-2 0 2 0

4 8 12

Difference in proportion choices-0.3 0 0 2 4 6 0.3 Session Pairs (#) Session Pairs (#)

Main experiment inactivations All inactivations

Degrees relative to target

Baseline

Inac

tiv

ation

Contralateral

choices Ipsilateralchoices

Degrees relative to target

-5 0 5 -5 0 5 31% of choices 56% of choices Example Session saccade target

Figure 3.7: Performance in the free-choice task following LIP inactivation. A, The effect of LIP inactivation on choice bias and saccade accuracy in the free-choice task, example session. Four panels show data from an ex- ample baseline/inactivation pair: saccade landing points (black dots) have been aligned to target position (red dot) for contralateral (left) and ipsilateral target choices (right), during baseline (top) and inactivation (bottom). Per- cent contralateral choices within a session are noted as text in the top left. B, The effect of LIP inactivation on choice bias and saccade accuracy in the free-choice task, over all sessions. Histograms show baseline/inactivation dif- ferences in proportion contralateral choices (top) and saccade error (bottom), where positive numbers indicate an increase in metric following inactivation. Dark bars indicate sessions that took place on the same days as the main direc- tion discrimination experiment (“Main experiment inactivations”, n=21); dark triangle indicates the median difference. Light bars include additional sessions that took place on other days (“All inactivations”, n=34); light triangle indi- cates median difference (may be hard to discern on plot due to similarity in value).

No systematic change was detected in other free-choice oculomotor met- rics (reaction time, peak velocity, or duration), and no change in any oculo- motor metric was detected during the direction discrimination task. Despite observing a muscimol-induced effect in the free-choice task, effect magnitude in the free-choice task was not predictive of effect magnitude in the direction discrimination task (Fig. 4.4 A, B), nor was there a dose-response relationship between muscimol mass and behavioral performance (Fig. 4.4 C-E), suggest- ing that our large muscimol administrations were likely operating within a “ceiling” regime.

3.3.4 Compensation over time or between hemispheres is unlikely Because muscimol inactivations require comparisons across relatively long time scales, it remains logically possible that LIP normally plays a critical role in decision-making, but that other areas are processing information in parallel (de Lafuente et al. 2015) and are able to quickly compensate when it is artificially inactivated. Although other techniques with faster time scales will allow for more direct tests of this possibility, we did not observe changes indicative of compensation either within a session (Fig. 4.5) or over sessions (Fig. 4.6).

To test for reliance or compensation involving the LIP in the non- inactivated hemisphere, we performed experiments with both choice targets placed within the contralateral hemifield, and again did not observe clear changes in behavioral performance (Fig. 3.7, A). We also found no disrup-

-3 0 3 Motion Strength (z) 0 0.5 1 Proportion Choices -3 0 3 Motion Strength (z) 0 0.5 1

Proportion Choices Baseline Inactivation

a

N = 6, 2 subjects N = 2, 1 subject

LIP

Both targets in inactivated field Newsome dots

b

LIP

Baseline Inactivation

Figure 3.8: A, Psychophysical data for pairs of baseline and muscimol treat- ment in LIP, when both choice targets were placed within the inactivated field. Inset presents stimulus geometry and estimated inactivated field. B,, Same format as A, for data collected when the motion stimulus was a random dot kinetogram (“Newsome dots”, at motion strengths of 0, 3.2, 6.4, 12.8 25.6 and 51.2% coherence, z scored).

tion of choice behavior using a moving-dot stimulus identical to that used in the classical studies of LIP function during decision making (Newsome & Pare 1988, Shadlen & Newsome 2001) (Fig. 3.8, B)

3.4

Discussion

Our results reveal a dissociation between decision-related activity in LIP and the causal role of such signals in decision-making. Instead, decision- related signals in LIP may be a result of feedback signal flow (Crowe et al. 2013), or perhaps an emergent phenomenon driven by extensive training (Sarma et al. 2015). Although one prior study observed subtle effects of LIP micros- timulation in a reaction time direction discrimination task (Hanks et al. 2006), such electrical perturbations can produce orthodromic (and antidromic) acti- vation of connected areas, and their observed effects are reconcilable with multiple alternatives to evidence accumulation (Hanks et al. 2015).

Alternatively, it remains possible that LIP does contribute to decision- making, but does so in a nonessential manner in conjunction with associated brain regions. Indeed, a growing body of work has observed decision-related activity in other brain areas (Ding & Gold 2013, Gu et al. 2007, Nienborg & Cumming 2009, Liu et al. 2013, Hanks et al. 2015), consistent with the prospect of LIP playing a minor and/or nonessential role in decision-making. In fact, our results mirror findings made in rodent posterior parietal cortex, where despite electrophysiological correlates of evidence accumulation, inactivation did not yield clear evidence of a critical role (Erlich et al. 2015). Taken to- gether, decision-related activity is likely represented broadly across the brain, and may be “read out” by a flexible process to support behavior (Pitkow et al. 2015, Raposo et al. 2014, Siegel et al. 2015). Our results call for a broader consideration of both decision-making circuitry and the mechanisms

for reading out decision-related activity— regardless of whether decisions are instantiated, or merely reflected, in a particular brain area.

Chapter 4

Extended analysis of Chapter 3:

Dissociated functional significance of

decision-related activity across the primate

dorsal stream

In Chapter 3 I reported results of an experiment testing whether decision- related activity plays a causal role in perceptual decision-making (under re- view, Katz LN, Yates JL, Pillow JW, Huk AC). In the experiment we found that reversible inactivation of macaque LIP, an area well known for its decision- related activity, has no impact on decision-making performance despite having exerted effects in a control task. This result stands to overturn a long held view and must therefore be scrutinized for its validity. Here I present fur- ther analyses, address specific concerns, and discuss the relation of this result to previous work. Parts of this chapter, with some modification, have been submitted as supplementary material to the results in Chapter 3.