Perceptual decision-making is a deliberative process that produces a categorical judgment regarding the presence, identity, and other features of a sensory stimulus (Gold and Shadlen, 2007). This deliberation often requires resolving potentially ambiguous interpretations of the current sensory stimulus with expectations that can be learned by evaluating prior decisions and their outcomes (Fecteau and Munoz, 2003; Gold and Stocker, 2017). This learning process can result in sequential effects on a subject’s choices and response times (RTs) when they participate in psychophysical tasks that require one decision after another (Gold et al., 2008; Marcos et al., 2013; Akaishi et al., 2014; Hwang et al., 2017; Akrami et al., 2018; Busse et al., 2011; Fischer and Whitney, 2014; Abrahamyan et al., 2016; Luu and Stocker, 2018). Because sequential effects can be present even when a task is designed to generate independent trial-by-trial choices and after extensive training, these effects may represent fundamental, ongoing processes that evaluate and adjust decisions to account for changes in the sensory environment, reward contingencies, and other factors (Fan et al., 2018; Gold and Stocker, 2017). Although neuronal substrates of these sequential effects have been identified in several brain regions (Barraclough et al., 2004; Ding and Gold, 2010; Ding and Gold, 2012a; Carnevale et al., 2012; Fecteau and Munoz, 2003; Marcos et al., 2013; Hwang et al., 2017; Akrami et al., 2018; Gold et al., 2008; Akaishi et al., 2014; St John-Saaltink et al., 2016; Lueckmann et al., 2018; Histed et al., 2009), the exact nature of the signals that the brain uses to support these effects is still not well understood.

In a previous study, we demonstrated that neurons in middle-lateral (ML) and anterolateral (AL) belt regions of the auditory cortex encode key features of the sensory evidence needed to solve an auditory-decision task (Tsunada et al., 2016). This task required monkeys to report whether a tone-burst sequence contained more low- or high- frequency tone bursts. We found that both AL and ML neurons were modulated by the frequency content of the tone-burst sequence. Moreover, in AL, but not ML, neuronal activity was weakly related to choice, and microstimulation biased the monkeys’ choices. These findings suggest a more direct role for AL in the decision process than for ML. They are also consistent with the idea that AL provides evidence for the decision but leave open the question of where and how in the brain is this evidence interpreted and combined with other information to form the decision (Gold and Shadlen, 2007).

The goal of the present study was to identify a role for the ventrolateral prefrontal cortex (vlPFC) in forming this auditory decision. We targeted vlPFC because it receives direct and indirect projections from AL and is situated at the apex of the ventral auditory pathway, which is commonly thought to mediate auditory perception and decision-making (Romanski et al., 1999; Hackett et al., 1999; Rauschecker and Tian, 2000; Russ et al., 2008b; Tsunada et al., 2016; Bizley and Cohen, 2013). Here we show that, contrary to our initial expectations, vlPFC neurons do not appear to encode information relevant to forming the current decision. Instead, vlPFC population activity can encode rich, graded information about the just-completed decision process, including the strength of the sensory evidence, the resulting choice, and whether or not it was correct. These signals, which were apparent from just after the decision was formed until after feedback was received, were closely and causally related to the subsequent decision in a manner that matched each monkey’s idiosyncratic choice strategy. Together, these results imply a role for the vlPFC in the ongoing evaluation and adjustment of auditory decisions.

We recorded and manipulated vlPFC spiking activity in two monkeys while they reported whether a noisy auditory stimulus contained more low- or high-frequency tone bursts (Figure 1a,b). A primary benefit of this task is that we could control the strength of the sensory evidence (the fraction of low or high tone bursts in a given stimulus; that is coherence) and relate that evidence to the monkeys’ choices and to vlPFC activity. These monkeys participated in our previous study (Tsunada et al., 2016), but the behavioral and neuronal data presented here have not been reported previously.

Figure 1

Download asset Open asset

Idiosyncratic choice-bias behavior

Both monkeys’ choice accuracy and RTs depended systematically on stimulus coherence (monkey T: n = 29 behavioral sessions; monkey A: n = 39 behavioral sessions; Figure 2a,b). For high-coherence stimuli, both monkeys almost always reported the correct answer with relatively short RTs. As absolute coherence decreased, performance accuracy decreased and RT increased. These psychometric (choice) and chronometric (RT) data were well described jointly by a drift-diffusion model (DDM; Figure 2a,b, pink lines; Gold and Shadlen, 2007; Ratcliff and McKoon, 2008). The DDM describes the process of forming a decision by accumulating incoming auditory (sensory) evidence over time to one of two pre-defined boundaries and accounts for both the choice (which boundary was reached) and the decision time (when the boundary was reached). These fits provided a consistently good match to the data (average deviance was 11.82 for monkey T [χ² -cumulative distribution, p<0.001] and 10.99 for monkey A [p<0.001]; Wichmann and Hill, 2001). This result indicates that the monkeys had consistent decision strategies, which were independent of the slight session-by-session differences in the choice of low- and high-frequency stimuli.

Figure 2

Download asset Open asset

Moreover, these fits partitioned the monkeys’ RTs into decision and non-decision times (Figure 2a,b), which facilitated our ability to identify the contributions of vlPFC activity to decision-making (Cohen and Newsome, 2009). In general, decision times for both choices increased as the absolute stimulus coherence decreased, which is a typical feature of the DDM (Gold and Shadlen, 2007). In contrast, non-decision times tended to be strongly asymmetric for the two choices, which reflected differences in the speed and preparation time of the right versus left joystick movements produced by the two monkeys. In subsequent analyses, we define the time of the ‘decision commitment’ as the end of the decision time plus an additional 50 ms to account for stimulus encoding (Tsunada et al., 2016).

We also identified idiosyncratic sequential choice biases for the two monkeys. Monkey T tended to use a ‘win-stay, lose-switch’ strategy (top panels in Figure 2c,d). That is, this monkey tended to repeat the previous choice if that choice was rewarded but switched choices if the previous choice was not rewarded. This tendency was seen as stay-switch biases with positive values following rewarded trials and negative values following non-rewarded trials for both the pooled data (Figure 2c; logistic regression: β₀ [high/low bias]=−0.46, p<0.01; β₁ [stimulus coherence]=2.50, p<0.01; β₂ [stay/switch bias given that the previous trial was rewarded]=0.14, p<0.01; β₃ [stay/switch bias given that the previous trial was not rewarded]=−0.30, p<0.01) and for the session-by-session data (Figure 2d; median β₀ = −0.50, Wilcoxon sign-rank test, p=0.04; β₁ = 3.73, p<0.01; β₂ = 0.27, p<0.01; β₃ = −0.39, p=0.03).

In contrast, monkey A tended to use a ‘win-switch’ strategy (bottom panels in Figure 2c,d). That is, this monkey tended to switch choices following a rewarded choice. Once again, this result was evident in both the pooled data (Figure 2c, β₀ = −0.45, p<0.01; β₁ = 3.16, p<0.01; β₂ = −0.19, p<0.01; β₃ = −0.02, p=0.88) and in the session-by-session data (Figure 2d, β₀ = 0.50, p<0.01; β₁ = 3.63, p<0.01; β₂ = −0.19, p<0.01; β₃ = 0.17, p=0.63). For both monkeys, we could not identify similar systematic effects on sequential RTs.

Post-decision neuronal representations of choice, outcome, and stimulus strength

Identifying the neuronal substrates of a perceptual decision typically involves identifying at least three forms of selectivity: (1) for choice, reflecting the consequence of the decision process; (2) for whether the choice was correct or an error, reflecting a closer association with perception than just sensory or motor processing; and (3) for stimulus strength because the process of forming the decision should reflect not just the categorical choice but also the strength of the evidence used to arrive at that choice (Gold and Shadlen, 2007). As detailed below, we identified all three forms of selectivity in vlPFC activity but only after the time of decision commitment on the current trial.

Individual vlPFC neurons had task-driven activity that was modulated selectively by the monkeys’ low- versus high-frequency choices (single-unit examples are shown in Figure 3 and summary population data are shown in Figure 4). Across the population of recorded neurons, the onset of choice selectivity for individual neurons (which were recorded in separate sessions and across the two monkeys) spanned the time from the inferred decision commitment through the motor response (joystick movement) and to the time when the reward was delivered or withheld. This selectivity included preferences for both high- and low-frequency choices (corresponding to ipsilateral and contralateral choices, respectively, because high-/low- frequency choices were indicated with leftward/rightward movements, and we recorded from the left hemisphere in both monkeys). We identified 13 and 15 high-frequency (ipsilateral)-preferring neurons in monkeys T and A, respectively; and 10 and 14 low-frequency (contralateral)-preferring neurons, respectively (there was not any evidence for laterality: χ²-test for H₀: no difference in the proportion of contralateral- and ipsilateral preferring neurons, p>0.05 for both monkeys).

Figure 3

Download asset Open asset

Figure 4

Download asset Open asset

The choice selectivity of individual neurons was also affected by the outcome of the current trial. Specifically, choice selectivity tended to be higher on correct trials than on error trials (Figure 5). This difference in choice selectivity was apparent even before reward delivery and thus could not be explained trivially as a direct response to reward delivery. Instead, this difference likely reflected differences in decision processing on correct versus error trials, including more uncertainty in the neuronal representation of the sensory evidence and therefore possibly lower reward expectations on error trials (Gold and Shadlen, 2003).

Figure 5

Download asset Open asset

In contrast to these single-neuron modulations by choice and outcome, the responses of individual vlPFC neurons were not selective for stimulus coherence. In particular, we found that, at any given time point, at most six vlPFC neurons were modulated by stimulus coherence for either preferred or non-preferred choices, which is not more than would be expected by chance in our sample (Figure 4b). However, we found a more robust representation of stimulus coherence at the level of population neuronal activity. The intuition for this discrepancy can be seen in Figure 4a: the fraction of neurons with choice selectivity was systematically smaller with decreasing coherence, implying that population-level signals were dependent on coherence. Although these effects did not correspond to statistically reliable differences in single-neuron coherence selectivity, there was enough information across the population of 103 neurons recorded from all sessions from both monkeys for a linear classifier to decode both stimulus coherence and choice (Figure 6). In both cases, decoding accuracy rose above chance levels only after the end of the decision process (blue vertical stripes in the stimulus-aligned plots in the left panels of Figure 6) and peaked around the time reward delivery, implying a form of post-decision processing.

Figure 6

Download asset Open asset

Thus, the vlPFC population had access to the key features of the decision process, including the strength of the evidence used to form the decision, the choice, and whether the choice was correct or an error. These signals were not apparent before the decision commitment, implying that they did not contribute to the formation of the current decision. Instead, the timing of these signals suggests that they may play a role in post-decision processing that can link one decision to the next decision.

Post-decision vlPFC activity encodes subsequent choice biases

Neuronal activity in the post-decision epoch was selective for both the current choice and the subsequent choice. This selectivity for the subsequent choice also matched each monkey’s idiosyncratic choice biases (Figure 2c,d), particularly following rewarded trials. Specifically, monkey T’s tendency to repeat rewarded trials was reflected in post-decision neuronal responses that tended to be slightly larger on trials in which the subsequent choice matched the neuron’s choice selectivity (Figure 7 top). For example, if a neuron tended to respond more for a high-frequency choice in the post-decision epoch of the current trial, its response tended, on average, to be slightly higher when the monkey made a high-frequency choice on the subsequent trial. In contrast, monkey A’s tendency to switch after rewarded trials was reflected in neuronal responses that tended, on average, to be slightly smaller on trials in which the subsequent choice matched the neuron’s choice selectivity (Figure 7 bottom). These effects had slightly different time courses in the two monkeys. Nonetheless, in both cases, these effects occurred after the decision commitment on correct trials, which corresponds to a time period when evaluative processing could be used to adjust subsequent decisions. We could not identify similarly reliable effects following error trials, possibly reflecting the much smaller data sets from those trials.

Figure 7

Download asset Open asset

Electrical microstimulation biased the subsequent choice

We used electrical microstimulation to test whether vlPFC activity plays a causal role in driving choice biases on the subsequent trial. We applied microstimulation from the time of stimulus onset until just after the behavioral response on a randomly selected 50% of trials in a subset of sessions (n = 11 sessions for monkey T, 21 sessions for monkey A). Despite the fact that this protocol was designed to test our initial hypothesis that vlPFC activity encoded formation of the current decision (and thus microstimulation was applied primarily during decision formation), we found that microstimulation did not systematically affect either the choice bias or the sensitivity of the decision on the current trial (single-site examples are shown in Figure 8a,b; population summaries are shown in Figure 8c,d).

Figure 8

Download asset Open asset

Instead, we found that microstimulation induced choice biases for the subsequent decision that depended systematically on the choice selectivity of the recorded vlPFC neuron at the microstimulation site. If microstimulation was applied at a site that was tuned for low-frequency choices, it tended to cause a low-choice bias on the subsequent trial (single-site example in Figure 8e). In contrast, if microstimulation was applied at a site that was tuned for high-frequency choices, it tended to cause a high-choice bias on the subsequent trial (single-site example in Figure 8f). Accordingly, across the population of microstimulation sites from both monkeys, the induced choice bias on the subsequent trial was correlated positively with the strength and direction of choice selectivity at the given site (Figure 8g). In other words, microstimulation at a site with neuronal activity that was selective for a low- or high-frequency choice on the current trial biased the monkeys’ choices toward a low- or high-frequency choice, respectively, on the subsequent trial. Further, the absolute magnitude of this bias was positively correlated with the strength of choice selectivity at the site of microstimulation. We did not identify any concomitant, systematic changes in psychometric sensitivity (Figure 8h).

These microstimulation effects also depended on specific choice patterns, albeit slightly differently for the two monkeys. When microstimulation was applied on a trial that resulted in a rewarded high choice, the subsequent choice tended to be biased in the same direction as the choice selectivity of the neuron recorded at the site of microstimulation for both monkeys (monkey T: Spearman’s rank correlation coefficient ρ = 0.90, p<0.001; monkey A: ρ = 0.61, p=0.04). When microstimulation was applied on a trial that resulted in a rewarded low choice, a similar effect was found only for one of the two monkeys (monkey T: ρ = 0.91, p<0.001; monkey A: ρ = 0.10, p=0.75). Together, these effects are consistent with the hypothesis that vlPFC activity is causally involved in evaluative processing that adjusts subsequent decisions.

We combined behavioral modeling, neuronal recordings, and electrical microstimulation to identify causal contributions of the primate vlPFC to a simple auditory perceptual decision about the frequency content of a sequence of tone bursts. vlPFC population activity had many of the hallmarks of a decision variable that could account for the monkeys’ patterns of choices and RTs, including selectivity for the strength of the sensory evidence, the monkey’s choice, and whether the choice was correct or incorrect. However, these forms of selectivity were not evident until after the time of the decision commitment and thus were inconsistent with a role for these neurons in forming the current decision. Instead, this post-decision selectivity appeared to support sequential adjustments to the decision process. Specifically, post-decision neuronal activity was modulated by each monkey’s idiosyncratic sequential choice biases. Further, electrical microstimulation at vlPFC sites affected the monkeys’ choices on the subsequent, but not the current, trial. Together, these results imply a role for post-decision vlPFC activity in encoding information about the process of forming the just-completed decision that is used to generate individualized biases that affect the subsequent decision.

These post-decision signals are a form of ‘decision-trace’ activity that has been identified in numerous brain areas, including parts of the prefrontal and parietal cortices (Barraclough et al., 2004; Bizzi, 1968; Funahashi et al., 1991; Tsujimoto and Sawaguchi, 2004; Ding and Gold, 2012a; Hwang et al., 2017; Akrami et al., 2018; Ding and Gold, 2010; Histed et al., 2009). This kind of activity represents information about the immediately preceding decision that can be used as part of a feedback-driven learning process to adjust future decisions based on a comparison between the expected and actual outcome of prior decisions (Sutton and Barto, 1998). Consistent with this idea, the population-level post-decision representations that we identified in the vlPFC are, in principle, sufficient to compute a confidence or reward-expectation signal for the just-completed decision (Ding and Gold, 2010; Ding and Gold, 2012a; Kiani and Shadlen, 2009; Kepecs et al., 2008). This signal could then be compared to the actual outcome to adjust the subsequent decision, possibly in other cortical or subcortical brain regions known to encode reward feedback and prediction error (Schultz, 2015).

The fact that our two monkeys used different biasing strategies (win-stay for monkey T, win-switch for monkey A) affects the interpretation of our findings in two primary ways. First, vlPFC selectivity for the subsequent choice was consistent with each monkey’s strategy. This result provides stronger support for the behavioral relevance of these signals than if they encoded features of behavior that were only present on average. Second, our microstimulation effects on the subsequent choice tended to depend on the choice tuning of neurons at the site of microstimulation and not the idiosyncratic sequential bias strategy of the monkey. This result implies that the vlPFC provides a choice-dependent signal that is used to generate sequential biases but may not participate directly in forming the idiosyncratic strategies that use those biases.

These findings are broadly consistent with recent studies that have shown a role for the vlPFC in strategy-switching and probabilistic-learning tasks, which share features of the sequential effects that we identified (Baxter et al., 2009; Rudebeck et al., 2017). However, our findings of primarily post-decision processing in the vlPFC are somewhat inconsistent with other studies that have implicated the PFC in forming auditory and other decisions (Russ et al., 2008b; Cohen et al., 2009; Lee et al., 2009; Bizley and Cohen, 2013; Kim and Shadlen, 1999). The reasons for this discrepancy are not clear. One possible reason is that we sampled a different PFC population than in these other studies, and these different populations make different contributions to decision versus post-decision processing. Another possibility is that previous reports of decision-related activity were also largely post-decisional. However, accounting for this possibility is not straightforward: the previous studies did not use RT tasks, making it difficult to interpret those results in terms of whether the reported decision-related signals occurred before or after the decision was formed on each trial (Russ et al., 2008b; Cohen et al., 2009; Fritz et al., 2010).

We still do not know the brain regions that form the decision for our task and consequently do not understand the mechanisms underlying these decisions. Because the monkeys’ choice and RT patterns reflected both the temporal accumulation of sensory evidence and sequential choice biases, we would expect these putative brain regions to implement two key operations. First, they should temporally accumulate the sensory evidence that is represented in the auditory cortex, particularly AL (Tsunada et al., 2016). Second, they should combine this accumulated evidence with historical sensory, outcome, and choice information, similar to that represented in vlPFC, to drive sequential choice biases (Ding and Gold, 2012a; Ding and Gold, 2010; Kim and Shadlen, 1999; Hwang et al., 2017; Akrami et al., 2018). One possibility is a set of other brain areas that have shown information-accumulation activity on other tasks, such as the dorsolateral PFC and parts of the posterior parietal cortex (Gold and Shadlen, 2007; Brody and Hanks, 2016). The posterior parietal cortex is a particularly compelling target of future studies because in rodents, it contributes to history-dependent choice biases on an auditory decision task (Akrami et al., 2018). Another intriguing possibility is an auditory-specific circuit involving the parabelt region of auditory cortex, which receives direct input from AL, projects to the vlPFC, and analyzes acoustic properties of behaviorally relevant sounds (Hackett et al., 1999; Romanski et al., 1999; Petkov et al., 2004).

Future studies that aim to identify neuronal activity related to decision formation for this kind of auditory task will likely benefit from not only an RT design to better identify the temporal epoch of decision formation but also a more thorough understanding of the dynamic and possibly idiosyncratic nature of the computations used to convert incoming sensory evidence into the categorical choice (Cohen and Newsome, 2009; Fan et al., 2018). These studies might also benefit from analyses that focus on substrates of subject-specific decision strategies, which help to establish the behavioral relevance of the neuronal signals (Fan et al., 2018; Busse et al., 2011; Abrahamyan et al., 2016). For example, in the present study, we found that our monkeys had different strategies of sequential choice biases (win-stay for monkey T versus win-switch for monkey A; Figure 2). These kinds of subject-specific choice biases have been reported previously in humans and animals, but their neuronal correlates have yet to be fully explained (Busse et al., 2011; Abrahamyan et al., 2016; Ding and Gold, 2010). Because the monkeys’ idiosyncratic behavioral strategies corresponded to different patterns of vlPFC choice selectivity, it implies that these signals may play a behaviorally relevant, subject-specific role in the evaluation and adjustment of the decision process, rather than providing a simple memory trace of common components of the decision process (Tsujimoto and Postle, 2012; Tsujimoto and Sawaguchi, 2004).

It is worth emphasizing that in our study, the representations in the vlPFC of critical decision-related variables (i.e., choice, outcome, and the strength of the sensory evidence) were not all evident in the spiking activity of individual neurons but instead were seen at the level of neuronal populations. The time course of these representations was also distributed across the neuronal population: different neurons (in our case, recorded in separate sessions) responded selectively at relatively restricted times with onsets that tiled the task epoch in a manner similar to other reports of working memory in the prefrontal cortex (Zaksas and Pasternak, 2006; Jun et al., 2010; Lundqvist et al., 2016; Brody et al., 2003; Schmitt et al., 2017). These population-level representations highlight the importance of conducting population recordings and analyses to identify and understand complex decision-related computations in the brain (Murray et al., 2017; Pouget et al., 2000; Kohn et al., 2016; Averbeck et al., 2006; Meyers, 2018; Meister et al., 2013).

In addition to recording from large neuronal populations, it would be instructive to examine vlPFC activity under a broader range of task conditions to better understand its general role in the sequential processing of auditory information. Our findings suggest that vlPFC can support auditory processing across trials. Other studies have shown that vlPFC also can play a role in rule-based sequence learning, which requires complex temporal processing over multiple time scales (Wilson et al., 2015). This kind of flexible temporal processing has been associated with human syntactic processing, a key feature of language (Wilson et al., 2015; Kikuchi et al., 2017; Wilson et al., 2017). It would be interesting to explore whether monkey vlPFC may possess a precursor system used in human language.

The University of Pennsylvania Institutional Animal Care and Use Committee approved all of the experimental protocols, which were conducted under protocol 804699. All surgical procedures were conducted using aseptic surgical techniques and with the monkeys kept under general anesthesia. A transparent reporting form is available. The authors were not blind to group allocation during the experiment and when assessing the data outcomes.

Two male monkeys (Macaca mulatta; monkey T [15 years old] and monkey A [14 years old]) participated in this study. Both were used in a previous study of auditory cortex (Tsunada et al., 2016), and monkey T was also used in a previous study of vlPFC (Tsunada et al., 2011a). In each session, the monkey was seated in a primate chair. A calibrated speaker (model MSP7, Yamaha) was placed in front of the monkey at eye level. The monkey moved a joystick, which was attached to the primate chair, to indicate their behavioral report. All experimental sessions took place in an RF-shielded room that had sound-attenuating walls and echo-absorbing foam on the inner walls.

The data analyses were performed in Matlab; this code is available https://github.com/CohenAuditoryLab/Joji (copy archived at https://github.com/elifesciences-publications/Joji).

1. 1. Abrahamyan A
   2. Silva LL
   3. Dakin SC
   4. Carandini M
   5. Gardner JL

   (2016) Adaptable history biases in human perceptual decisions

   PNAS 113:E3548–E3557.

   https://doi.org/10.1073/pnas.1518786113

   • PubMed
   • Google Scholar
2. 1. Akaishi R
   2. Umeda K
   3. Nagase A
   4. Sakai K

   (2014) Autonomous mechanism of internal choice estimate underlies decision inertia

   Neuron 81:195–206.

   https://doi.org/10.1016/j.neuron.2013.10.018

   • PubMed
   • Google Scholar
3. 1. Akrami A
   2. Kopec CD
   3. Diamond ME
   4. Brody CD

   (2018) Posterior parietal cortex represents sensory history and mediates its effects on behaviour

   Nature 554:368–372.

   https://doi.org/10.1038/nature25510

   • PubMed
   • Google Scholar
4. 1. Averbeck BB
   2. Latham PE
   3. Pouget A

   (2006) Neural correlations, population coding and computation

   Nature Reviews Neuroscience 7:358–366.

   https://doi.org/10.1038/nrn1888

   • PubMed
   • Google Scholar
5. 1. Barraclough DJ
   2. Conroy ML
   3. Lee D

   (2004) Prefrontal cortex and decision making in a mixed-strategy game

   Nature Neuroscience 7:404–410.

   https://doi.org/10.1038/nn1209

   • PubMed
   • Google Scholar
6. 1. Baxter MG
   2. Gaffan D
   3. Kyriazis DA
   4. Mitchell AS

   (2009) Ventrolateral prefrontal cortex is required for performance of a strategy implementation task but not reinforcer devaluation effects in rhesus monkeys

   European Journal of Neuroscience 29:2049–2059.

   https://doi.org/10.1111/j.1460-9568.2009.06740.x

   • PubMed
   • Google Scholar
7. Book

   1. Bishop CM

   (2006)

   Pattern Recognition and Machine Learning

   Springer.

   • Google Scholar
8. 1. Bizley JK
   2. Cohen YE

   (2013) The what, where and how of auditory-object perception

   Nature Reviews Neuroscience 14:693–707.

   https://doi.org/10.1038/nrn3565

   • PubMed
   • Google Scholar
9. 1. Bizzi E

   (1968) Discharge of frontal eye field neurons during saccadic and following eye movements in unanesthetized monkeys

   Experimental Brain Research 6:69–80.

   https://doi.org/10.1007/BF00235447

   • PubMed
   • Google Scholar
10. 1. Brody CD
    2. Hernández A
    3. Zainos A
    4. Romo R

    (2003) Timing and neural encoding of somatosensory parametric working memory in macaque prefrontal cortex

    Cerebral Cortex 13:1196–1207.

    https://doi.org/10.1093/cercor/bhg100

    • PubMed
    • Google Scholar
11. 1. Brody CD
    2. Hanks TD

    (2016) Neural underpinnings of the evidence accumulator

    Current Opinion in Neurobiology 37:149–157.

    https://doi.org/10.1016/j.conb.2016.01.003

    • PubMed
    • Google Scholar
12. 1. Brunton BW
    2. Botvinick MM
    3. Brody CD

    (2013) Rats and humans can optimally accumulate evidence for decision-making

    Science 340:95–98.

    https://doi.org/10.1126/science.1233912

    • PubMed
    • Google Scholar
13. 1. Busse L
    2. Ayaz A
    3. Dhruv NT
    4. Katzner S
    5. Saleem AB
    6. Schölvinck ML
    7. Zaharia AD
    8. Carandini M

    (2011) The detection of visual contrast in the behaving mouse

    Journal of Neuroscience 31:11351–11361.

    https://doi.org/10.1523/JNEUROSCI.6689-10.2011

    • PubMed
    • Google Scholar
14. 1. Carnevale F
    2. de Lafuente V
    3. Romo R
    4. Parga N

    (2012) Internal signal correlates neural populations and biases perceptual decision reports

    PNAS 109:18938–18943.

    https://doi.org/10.1073/pnas.1216799109

    • PubMed
    • Google Scholar
15. 1. Cohen YE
    2. Russ BE
    3. Davis SJ
    4. Baker AE
    5. Ackelson AL
    6. Nitecki R

    (2009) A functional role for the ventrolateral prefrontal cortex in non-spatial auditory cognition

    PNAS 106:20045–20050.

    https://doi.org/10.1073/pnas.0907248106

    • PubMed
    • Google Scholar
16. 1. Cohen MR
    2. Newsome WT

    (2009) Estimates of the contribution of single neurons to perception depend on timescale and noise correlation

    Journal of Neuroscience 29:6635–6648.

    https://doi.org/10.1523/JNEUROSCI.5179-08.2009

    • PubMed
    • Google Scholar
17. Book

    1. Cox DR

    (1970)

    Analysis of binary data

    London: Methuen.

    • Google Scholar
18. 1. Ding L
    2. Gold JI

    (2010) Caudate encodes multiple computations for perceptual decisions

    Journal of Neuroscience 30:15747–15759.

    https://doi.org/10.1523/JNEUROSCI.2894-10.2010

    • PubMed
    • Google Scholar
19. 1. Ding L
    2. Gold JI

    (2012a) Neural correlates of perceptual decision making before, during, and after decision commitment in monkey frontal eye field

    Cerebral Cortex 22:1052–1067.

    https://doi.org/10.1093/cercor/bhr178

    • PubMed
    • Google Scholar
20. 1. Ding L
    2. Gold JI

    (2012b) Separate, causal roles of the caudate in Saccadic choice and execution in a perceptual decision task

    Neuron 75:865–874.

    https://doi.org/10.1016/j.neuron.2012.07.021

    • PubMed
    • Google Scholar
21. 1. Eckhoff P
    2. Holmes P
    3. Law C
    4. Connolly PM
    5. Gold JI

    (2008) On diffusion processes with variable drift rates as models for decision making during learning

    New Journal of Physics 10:015006.

    https://doi.org/10.1088/1367-2630/10/1/015006

    • PubMed
    • Google Scholar
22. 1. Fan Y
    2. Gold JI
    3. Ding L

    (2018) Ongoing, rational calibration of reward-driven perceptual biases

    eLife 7:e36018.

    https://doi.org/10.7554/eLife.36018

    • PubMed
    • Google Scholar
23. 1. Fecteau JH
    2. Munoz DP

    (2003) Exploring the consequences of the previous trial

    Nature Reviews Neuroscience 4:435–443.

    https://doi.org/10.1038/nrn1114

    • PubMed
    • Google Scholar
24. 1. Fischer J
    2. Whitney D

    (2014) Serial dependence in visual perception

    Nature Neuroscience 17:738–743.

    https://doi.org/10.1038/nn.3689

    • PubMed
    • Google Scholar
25. 1. Frey S
    2. Comeau R
    3. Hynes B
    4. Mackey S
    5. Petrides M

    (2004) Frameless stereotaxy in the nonhuman primate

    NeuroImage 23:1226–1234.

    https://doi.org/10.1016/j.neuroimage.2004.07.001

    • PubMed
    • Google Scholar
26. 1. Fritz JB
    2. David SV
    3. Radtke-Schuller S
    4. Yin P
    5. Shamma SA

    (2010) Adaptive, behaviorally gated, persistent encoding of task-relevant auditory information in ferret frontal cortex

    Nature Neuroscience 13:1011–1019.

    https://doi.org/10.1038/nn.2598

    • PubMed
    • Google Scholar
27. 1. Funahashi S
    2. Bruce CJ
    3. Goldman-Rakic PS

    (1991) Neuronal activity related to saccadic eye movements in the monkey's dorsolateral prefrontal cortex

    Journal of Neurophysiology 65:1464–1483.

    https://doi.org/10.1152/jn.1991.65.6.1464

    • PubMed
    • Google Scholar
28. 1. Gold JI
    2. Law CT
    3. Connolly P
    4. Bennur S

    (2008) The relative influences of priors and sensory evidence on an oculomotor decision variable during perceptual learning

    Journal of Neurophysiology 100:2653–2668.

    https://doi.org/10.1152/jn.90629.2008

    • PubMed
    • Google Scholar
29. 1. Gold JI
    2. Shadlen MN

    (2003) The influence of behavioral context on the representation of a perceptual decision in developing oculomotor commands

    The Journal of Neuroscience 23:632–651.

    https://doi.org/10.1523/JNEUROSCI.23-02-00632.2003

    • PubMed
    • Google Scholar
30. 1. Gold JI
    2. Shadlen MN

    (2007) The neural basis of decision making

    Annual Review of Neuroscience 30:535–574.

    https://doi.org/10.1146/annurev.neuro.29.051605.113038

    • PubMed
    • Google Scholar
31. 1. Gold JI
    2. Stocker AA

    (2017) Visual Decision-Making in an uncertain and dynamic world

    Annual Review of Vision Science 3:227–250.

    https://doi.org/10.1146/annurev-vision-111815-114511

    • PubMed
    • Google Scholar
32. 1. Green CS
    2. Pouget A
    3. Bavelier D

    (2010) Improved probabilistic inference as a general learning mechanism with action video games

    Current Biology 20:1573–1579.

    https://doi.org/10.1016/j.cub.2010.07.040

    • PubMed
    • Google Scholar
33. 1. Hackett TA
    2. Stepniewska I
    3. Kaas JH

    (1999) Prefrontal connections of the parabelt auditory cortex in macaque monkeys

    Brain Research 817:45–58.

    https://doi.org/10.1016/S0006-8993(98)01182-2

    • PubMed
    • Google Scholar
34. 1. Hanks TD
    2. Ditterich J
    3. Shadlen MN

    (2006) Microstimulation of macaque area LIP affects decision-making in a motion discrimination task

    Nature Neuroscience 9:682–689.

    https://doi.org/10.1038/nn1683

    • PubMed
    • Google Scholar
35. 1. Histed MH
    2. Pasupathy A
    3. Miller EK

    (2009) Learning substrates in the primate prefrontal cortex and striatum: sustained activity related to successful actions

    Neuron 63:244–253.

    https://doi.org/10.1016/j.neuron.2009.06.019

    • PubMed
    • Google Scholar
36. 1. Hwang EJ
    2. Dahlen JE
    3. Mukundan M
    4. Komiyama T

    (2017) History-based action selection bias in posterior parietal cortex

    Nature Communications 8:1242.

    https://doi.org/10.1038/s41467-017-01356-z

    • PubMed
    • Google Scholar
37. 1. Johnston JM
    2. Cohen YE
    3. Shirley H
    4. Tsunada J
    5. Bennur S
    6. Christison-Lagay K
    7. Veeder CL

    (2016) Recent refinements to cranial implants for rhesus macaques (Macaca mulatta)

    Lab Animal 45:180–186.

    https://doi.org/10.1038/laban.997

    • PubMed
    • Google Scholar
38. 1. Jun JK
    2. Miller P
    3. Hernández A
    4. Zainos A
    5. Lemus L
    6. Brody CD
    7. Romo R

    (2010) Heterogenous population coding of a short-term memory and decision task

    Journal of Neuroscience 30:916–929.

    https://doi.org/10.1523/JNEUROSCI.2062-09.2010

    • PubMed
    • Google Scholar
39. 1. Kepecs A
    2. Uchida N
    3. Zariwala HA
    4. Mainen ZF

    (2008) Neural correlates, computation and behavioural impact of decision confidence

    Nature 455:227–231.

    https://doi.org/10.1038/nature07200

    • PubMed
    • Google Scholar
40. 1. Kiani R
    2. Shadlen MN

    (2009) Representation of confidence associated with a decision by neurons in the parietal cortex

    Science 324:759–764.

    https://doi.org/10.1126/science.1169405

    • PubMed
    • Google Scholar
41. 1. Kikuchi Y
    2. Attaheri A
    3. Wilson B
    4. Rhone AE
    5. Nourski KV
    6. Gander PE
    7. Kovach CK
    8. Kawasaki H
    9. Griffiths TD
    10. Howard MA
    11. Petkov CI

    (2017) Sequence learning modulates neural responses and oscillatory coupling in human and monkey auditory cortex

    PLOS Biology 15:e2000219.

    https://doi.org/10.1371/journal.pbio.2000219

    • PubMed
    • Google Scholar
42. 1. Kim JN
    2. Shadlen MN

    (1999) Neural correlates of a decision in the dorsolateral prefrontal cortex of the macaque

    Nature Neuroscience 2:176–185.

    https://doi.org/10.1038/5739

    • PubMed
    • Google Scholar
43. 1. Kohn A
    2. Coen-Cagli R
    3. Kanitscheider I
    4. Pouget A

    (2016) Correlations and neuronal population information

    Annual Review of Neuroscience 39:237–256.

    https://doi.org/10.1146/annurev-neuro-070815-013851

    • PubMed
    • Google Scholar
44. 1. Lee JH
    2. Russ BE
    3. Orr LE
    4. Cohen YE

    (2009) Prefrontal activity predicts monkeys' decisions during an auditory category task

    Frontiers in Integrative Neuroscience 3:.

    https://doi.org/10.3389/neuro.07.016.2009

    • PubMed
    • Google Scholar
45. 1. Lueckmann JM
    2. Macke JH
    3. Nienborg H

    (2018) Can serial dependencies in choices and neural activity explain choice probabilities?

    The Journal of Neuroscience 38:3495–3506.

    https://doi.org/10.1523/JNEUROSCI.2225-17.2018

    • PubMed
    • Google Scholar
46. 1. Lundqvist M
    2. Rose J
    3. Herman P
    4. Brincat SL
    5. Buschman TJ
    6. Miller EK

    (2016) Gamma and beta bursts underlie working memory

    Neuron 90:152–164.

    https://doi.org/10.1016/j.neuron.2016.02.028

    • PubMed
    • Google Scholar
47. 1. Luu L
    2. Stocker AA

    (2018) Post-decision biases reveal a self-consistency principle in perceptual inference

    eLife 7:e33334.

    https://doi.org/10.7554/eLife.33334

    • PubMed
    • Google Scholar
48. 1. Marcos E
    2. Pani P
    3. Brunamonti E
    4. Deco G
    5. Ferraina S
    6. Verschure P

    (2013) Neural variability in premotor cortex is modulated by trial history and predicts behavioral performance

    Neuron 78:249–255.

    https://doi.org/10.1016/j.neuron.2013.02.006

    • PubMed
    • Google Scholar
49. 1. Meister ML
    2. Hennig JA
    3. Huk AC

    (2013) Signal multiplexing and single-neuron computations in lateral intraparietal area during decision-making

    Journal of Neuroscience 33:2254–2267.

    https://doi.org/10.1523/JNEUROSCI.2984-12.2013

    • PubMed
    • Google Scholar
50. 1. Meyers EM
    2. Freedman DJ
    3. Kreiman G
    4. Miller EK
    5. Poggio T

    (2008) Dynamic population coding of category information in inferior temporal and prefrontal cortex

    Journal of Neurophysiology 100:1407–1419.

    https://doi.org/10.1152/jn.90248.2008

    • PubMed
    • Google Scholar
51. 1. Meyers EM

    (2018) Dynamic population coding and its relationship to working memory

    Journal of Neurophysiology 120:2260–2268.

    https://doi.org/10.1152/jn.00225.2018

    • PubMed
    • Google Scholar
52. 1. Mulder MJ
    2. Keuken MC
    3. van Maanen L
    4. Boekel W
    5. Forstmann BU
    6. Wagenmakers EJ

    (2013) The speed and accuracy of perceptual decisions in a random-tone pitch task

    Attention, Perception, & Psychophysics 75:1048–1058.

    https://doi.org/10.3758/s13414-013-0447-8

    • PubMed
    • Google Scholar
53. 1. Murray JD
    2. Bernacchia A
    3. Roy NA
    4. Constantinidis C
    5. Romo R
    6. Wang XJ

    (2017) Stable population coding for working memory coexists with heterogeneous neural dynamics in prefrontal cortex

    PNAS 114:394–399.

    https://doi.org/10.1073/pnas.1619449114

    • PubMed
    • Google Scholar
54. 1. Petkov CI
    2. Kang X
    3. Alho K
    4. Bertrand O
    5. Yund EW
    6. Woods DL

    (2004) Attentional modulation of human auditory cortex

    Nature Neuroscience 7:658–663.

    https://doi.org/10.1038/nn1256

    • PubMed
    • Google Scholar
55. 1. Pouget A
    2. Dayan P
    3. Zemel R

    (2000) Information processing with population codes

    Nature Reviews Neuroscience 1:125–132.

    https://doi.org/10.1038/35039062

    • PubMed
    • Google Scholar
56. 1. Ratcliff R
    2. Van Zandt T
    3. McKoon G

    (1999) Connectionist and diffusion models of reaction time

    Psychological Review 106:261–300.

    https://doi.org/10.1037/0033-295X.106.2.261

    • PubMed
    • Google Scholar
57. 1. Ratcliff R
    2. McKoon G

    (2008) The diffusion decision model: theory and data for two-choice decision tasks

    Neural Computation 20:873–922.

    https://doi.org/10.1162/neco.2008.12-06-420

    • PubMed
    • Google Scholar
58. 1. Rauschecker JP
    2. Tian B

    (2000) Mechanisms and streams for processing of "what" and "where" in auditory cortex

    PNAS 97:11800–11806.

    https://doi.org/10.1073/pnas.97.22.11800

    • PubMed
    • Google Scholar
59. 1. Roitman JD
    2. Shadlen MN

    (2002) Response of neurons in the lateral intraparietal area during a combined visual discrimination reaction time task

    The Journal of Neuroscience 22:9475–9489.

    https://doi.org/10.1523/JNEUROSCI.22-21-09475.2002

    • PubMed
    • Google Scholar
60. 1. Romanski LM
    2. Tian B
    3. Fritz J
    4. Mishkin M
    5. Goldman-Rakic PS
    6. Rauschecker JP

    (1999) Dual streams of auditory afferents target multiple domains in the primate prefrontal cortex

    Nature Neuroscience 2:1131–1136.

    https://doi.org/10.1038/16056

    • PubMed
    • Google Scholar
61. 1. Romanski LM
    2. Goldman-Rakic PS

    (2002) An auditory domain in primate prefrontal cortex

    Nature Neuroscience 5:15–16.

    https://doi.org/10.1038/nn781

    • PubMed
    • Google Scholar
62. 1. Rudebeck PH
    2. Saunders RC
    3. Lundgren DA
    4. Murray EA

    (2017) Specialized Representations of Value in the Orbital and Ventrolateral Prefrontal Cortex: Desirability versus Availability of Outcomes

    Neuron 95:1208–1220.

    https://doi.org/10.1016/j.neuron.2017.07.042

    • PubMed
    • Google Scholar
63. 1. Russ BE
    2. Ackelson AL
    3. Baker AE
    4. Cohen YE

    (2008a) Coding of auditory-stimulus identity in the auditory non-spatial processing stream

    Journal of Neurophysiology 99:87–95.

    https://doi.org/10.1152/jn.01069.2007

    • PubMed
    • Google Scholar
64. 1. Russ BE
    2. Orr LE
    3. Cohen YE

    (2008b) Prefrontal neurons predict choices during an auditory same-different task

    Current Biology 18:1483–1488.

    https://doi.org/10.1016/j.cub.2008.08.054

    • PubMed
    • Google Scholar
65. 1. Salzman CD
    2. Britten KH
    3. Newsome WT

    (1990) Cortical microstimulation influences perceptual judgements of motion direction

    Nature 346:174–177.

    https://doi.org/10.1038/346174a0

    • PubMed
    • Google Scholar
66. 1. Schmitt LI
    2. Wimmer RD
    3. Nakajima M
    4. Happ M
    5. Mofakham S
    6. Halassa MM

    (2017) Thalamic amplification of cortical connectivity sustains attentional control

    Nature 545:219–223.

    https://doi.org/10.1038/nature22073

    • PubMed
    • Google Scholar
67. 1. Schultz W

    (2015) Neuronal reward and decision signals: from theories to data

    Physiological Reviews 95:853–951.

    https://doi.org/10.1152/physrev.00023.2014

    • PubMed
    • Google Scholar
68. 1. Selezneva E
    2. Scheich H
    3. Brosch M

    (2006) Dual time scales for categorical decision making in auditory cortex

    Current Biology 16:2428–2433.

    https://doi.org/10.1016/j.cub.2006.10.027

    • PubMed
    • Google Scholar
69. 1. Shadlen MN
    2. Hanks TD
    3. Churchland A
    4. Kiani R
    5. Yang T

    (2006)

    Bayesian Brain: Probabilistic Approaches to Neural Coding

    The Speed and Accuracy of a Simple Perceptual Decision: A Mathematical Primer, Bayesian Brain: Probabilistic Approaches to Neural Coding, Cambridge, MA, MIT Press, 10.7551/mitpress/9780262042383.003.0010.

    • Google Scholar
70. 1. St John-Saaltink E
    2. Kok P
    3. Lau HC
    4. de Lange FP

    (2016) Serial dependence in perceptual decisions is reflected in activity patterns in primary visual cortex

    The Journal of Neuroscience 36:6186–6192.

    https://doi.org/10.1523/JNEUROSCI.4390-15.2016

    • PubMed
    • Google Scholar
71. Book

    1. Sutton RS
    2. Barto AG

    (1998)

    Reinforcement Learning: An Introduction

    Cambridge, Mass: MIT Press.

    • Google Scholar
72. 1. Tsujimoto S
    2. Postle BR

    (2012) The prefrontal cortex and oculomotor delayed response: a reconsideration of the "mnemonic scotoma"

    Journal of Cognitive Neuroscience 24:627–635.

    https://doi.org/10.1162/jocn_a_00171

    • PubMed
    • Google Scholar
73. 1. Tsujimoto S
    2. Sawaguchi T

    (2004) Neuronal representation of response-outcome in the primate prefrontal cortex

    Cerebral Cortex 14:47–55.

    https://doi.org/10.1093/cercor/bhg090

    • PubMed
    • Google Scholar
74. 1. Tsunada J
    2. Baker AE
    3. Christison-Lagay KL
    4. Davis SJ
    5. Cohen YE

    (2011a) Modulation of cross-frequency coupling by novel and repeated stimuli in the primate ventrolateral prefrontal cortex

    Frontiers in Psychology 2:.

    https://doi.org/10.3389/fpsyg.2011.00217

    • PubMed
    • Google Scholar
75. 1. Tsunada J
    2. Lee JH
    3. Cohen YE

    (2011b) Representation of speech categories in the primate auditory cortex

    Journal of Neurophysiology 105:2634–2646.

    https://doi.org/10.1152/jn.00037.2011

    • PubMed
    • Google Scholar
76. 1. Tsunada J
    2. Liu AS
    3. Gold JI
    4. Cohen YE

    (2016) Causal contribution of primate auditory cortex to auditory perceptual decision-making

    Nature Neuroscience 19:135–142.

    https://doi.org/10.1038/nn.4195

    • PubMed
    • Google Scholar
77. Software

    1. Tsunada J

    (2019) Joji, version 2a4ffb7

    Github.

    https://github.com/CohenAuditoryLab/Joji
78. 1. Wichmann FA
    2. Hill NJ

    (2001) The psychometric function: I. fitting, sampling, and goodness of fit

    Perception & Psychophysics 63:1293–1313.

    https://doi.org/10.3758/BF03194544

    • PubMed
    • Google Scholar
79. 1. Wilson B
    2. Kikuchi Y
    3. Sun L
    4. Hunter D
    5. Dick F
    6. Smith K
    7. Thiele A
    8. Griffiths TD
    9. Marslen-Wilson WD
    10. Petkov CI

    (2015) Auditory sequence processing reveals evolutionarily conserved regions of frontal cortex in macaques and humans

    Nature Communications 6:8901.

    https://doi.org/10.1038/ncomms9901

    • PubMed
    • Google Scholar
80. 1. Wilson B
    2. Marslen-Wilson WD
    3. Petkov CI

    (2017) Conserved sequence processing in primate frontal cortex

    Trends in Neurosciences 40:72–82.

    https://doi.org/10.1016/j.tins.2016.11.004

    • PubMed
    • Google Scholar
81. 1. Zaksas D
    2. Pasternak T

    (2006) Directional signals in the prefrontal cortex and in area MT during a working memory for visual motion task

    Journal of Neuroscience 26:11726–11742.

    https://doi.org/10.1523/JNEUROSCI.3420-06.2006

    • PubMed
    • Google Scholar

Author details

1. Joji Tsunada

   1. Department of Otorhinolaryngology, University of Pennsylvania, Philadelphia, United States
   2. Department of Veterinary Medicine, Faculty of Agriculture, Iwate University, Morioka, Japan

   Contribution

   Conceptualization, Data curation, Software, Investigation, Visualization, Methodology, Writing—original draft, Writing—review and editing

   Competing interests

   No competing interests declared

   "This ORCID iD identifies the author of this article:" 0000-0002-9125-7578

2. Yale Cohen

   1. Department of Otorhinolaryngology, University of Pennsylvania, Philadelphia, United States
   2. Department of Neuroscience, University of Pennsylvania, Philadelphia, United States
   3. Department of Bioengineering, University of Pennsylvania, Philadelphia, United States

   Contribution

   Conceptualization, Software, Formal analysis, Funding acquisition, Visualization, Methodology, Writing—review and editing

   Contributed equally with

   Joshua I Gold

   For correspondence

   ycohen@pennmedicine.upenn.edu

   Competing interests

   No competing interests declared

   "This ORCID iD identifies the author of this article:" 0000-0002-0830-5162

3. Joshua I Gold

   Department of Neuroscience, University of Pennsylvania, Philadelphia, United States

   Contribution

   Conceptualization, Software, Formal analysis, Funding acquisition, Visualization, Methodology, Writing—review and editing

   Contributed equally with

   Yale Cohen

   Competing interests

   Reviewing editor, eLife

   "This ORCID iD identifies the author of this article:" 0000-0002-6018-0483

• 2,918

  views

• 462

  downloads

• 35

  citations

Views, downloads and citations are aggregated across all versions of this paper published by eLife.