Perceptual Modulation of Motor—But Not Visual—Responses in the Frontal Eye Field during an Urgent-Decision Task

Abstract

Neuronal activity in the frontal eye field (FEF) ranges from purely motor (related to saccade production) to purely visual (related to stimulus presence). According to numerous studies, visual responses correlate strongly with early perceptual analysis of the visual scene, including the deployment of spatial attention, whereas motor responses do not. Thus, functionally, the consensus is that visually responsive FEF neurons select a target among visible objects, whereas motor-related neurons plan specific eye movements based on such earlier target selection. However, these conclusions are based on behavioral tasks that themselves promote a serial arrangement of perceptual analysis followed by motor planning. So, is the presumed functional hierarchy in FEF an intrinsic property of its circuitry or does it reflect just one possible mode of operation? We investigate this in monkeys performing a rapid-choice task in which, crucially, motor planning always starts ahead of task-critical perceptual analysis, and the two relevant spatial locations are equally informative and equally likely to be target or distracter. We find that the choice is instantiated in FEF as a competition between oculomotor plans, in agreement with model predictions. Notably, although perception strongly influences the motor neurons, it has little if any measurable impact on the visual cells; more generally, the more dominant the visual response, the weaker the perceptual modulation. The results indicate that, contrary to expectations, during rapid saccadic choices perceptual information may directly modulate ongoing saccadic plans, and this process is not contingent on prior selection of the saccadic goal by visually driven FEF responses.

Decoupling speed and accuracy in an urgent decision-making task reveals multiple contributions to their trade-off

“…Similar ideas have also been advocated by Cisek and colleagues based on recordings from premotor areas (Cisek and Kalaska, 2005; Pastor-Bernier and Cisek, 2011), giving rise to a powerful modeling framework, the “urgency-gating model” (Cisek, 2006; Cisek et al, 2009; Thura et al, 2012), that is similar in spirit to our accelerated race-to-threshold model (see Costello et al, 2013). …”

Comparing fixed and collapsing boundary versions of the diffusion model

“…The results of single-cell recording studies of monkeys performing saccade-to-target decision tasks have been interpreted as evidence for an ‘urgency signal.’ Consistent with this interpretation, the RT distributions of monkeys may be more symmetric than typical RT distributions produced by human subjects. However, numerous other researchers have continued to use models which do not include collapsing boundaries or an urgency signal (Boucher, Palmeri, Logan, & Schall, 2007; Costello, Zhu, Salinas, & Stanford, 2013; Ding & Gold, 2010, 2012; Purcell et al, 2010; Purcell, Schall, Logan, & Palmeri, 2012; Ramakrishnan & Murthy, 2013; Ramakrishnan et al, 2012; Salinas, Shankar, Costello, Zhu, & Stanford, 2010; Salinas & Stanford, 2013; Shankar et al, 2011; Stanford, Shankar, Massoglia, Costello, & Salinas, 2010) and other monkey studies have yielded more typical, positively-skewed RT distributions (Ratcliff, Hasegawa, Hasegawa, Smith, & Segraves, 2007), as well as those that were fit with ex-Gaussian distributions, (Camalier et al, 2007; Heitz & Schall, 2012, 2013; Middlebrooks & Schall, 2014). Additionally, while a standard diffusion process does not map directly onto neural firing rates, patterns of neural firing rates have been modeled using racing diffusion processes without any kind of urgency signal (Ratcliff et al, 2007).…”

Visuomotor Functions in the Frontal Lobe

“…The role of the FEF and the SC in controlling saccade initiation has been investigated thoroughly in monkeys performing a task that required them to interrupt saccade preparation on a random subset of trials in response to new stimuli. A mathematical race model that accounts for performance of this saccade-countermanding task provides unique theoretical leverage for distinguishing between neurons that can contribute to the control of saccade production and those that cannot , Paré & Hanes 2003, Murthy et al 2009, Costello et al 2013). …”

“…The target selection process can influence saccade trajectory (McPeek 2006), but the selection observed in visual neurons has been experimentally dissociated from saccade production (Sato & Schall 2003, Murthy et al 2009, Ramakrishnan et al 2012, Costello et al 2013, Lee et al 2012). Many researchers have described FEF contributions to guidance of covert attention ( Juan et al 2004, Zhou & Thompson 2009, Khayat et al 2009, Squire et al 2013, inspiring the claim that the FEF embodies a salience map (Thompson & Bichot 2005).…”