The what, where, and why of priority maps and their interactions with visual working memory

Abstract

Priority maps are winner-take-all neural mechanisms thought to guide the allocation of covert and overt attention. Here, we go beyond this standard definition and argue that priority maps play a much broader role in controlling goal-directed behavior. We start by defining what priority maps are and where they might be found in the brain; we then ask why they exist—the function that they serve. We propose that this function is to communicate a goal state to the different effector systems, thereby guiding behavior. Within this framework, we speculate on how priority maps interact with visual working memory and introduce our common source hypothesis, the suggestion that this goal state is maintained in visual working memory and used to construct all of the priority maps controlling the various motor systems. Finally, we look ahead and suggest questions about priority maps that should be asked next.

Active Inference: A Process Theory

“…This may sound like an overcomplete representation of policies; however, this sort of architecture is implicit in salience maps in the brain (Santangelo, 2015;Zelinsky & Bisley, 2015). This is because a salience map represents the value (e.g., epistemic value or Bayesian surprise) of all possible actions (e.g., saccadic eye movements), from which the best action is selected: see Mirza, Adams, Mathys, and Friston (2016) for a simulation of saccadic searches and scene construction using the current scheme.…”

Brain bases of the automaticity via visual search task

“…On the other hand, "oddball tasks", which involve the detection of a salient target (i.e., oddball) presented infrequently (e.g., 5-15%) in a stream of frequent standard stimuli, would be more suitable for studying the ventral stream functioning (Kim, 2013, meta-analysis). However, this distinction between the dorsal and ventral system must be nuanced and understood from an experimental perspective, because bottom-up and top-down systems are in constant interaction and influence each other Shomstein, 2012;Zelinsky and Bisley, 2015). Indeed, the visual search task makes it possible to study the different processes associated with selective attention (e.g., response preparation, orienting of attention, target selection and identification) and the priority maps which are the "locations" where the bottom-up and top-down information is processed and, thus, where ventral and dorsal stream information meet (Itti and Koch, 2001;Fecteau and Munoz, 2006;Serences and Yantis, 2006;Bisley, et al, 2009;Ipata et al, 2009;Zelinsky and Bisley, 2015;).…”

“…It is worth noting that, depending on the task is being studied, one could emphasize more the bottom-up aspects of the visual processing (i.e., free viewing tasks) or the top-down aspects (i.e., visual search tasks) because the characteristics of the tasks induce it (Zelinsky and Bisley, 2015). However, it is rather strange that prior knowledge does not influence what we are seeing or, in turn, that active guided visual scanning of the world is somewhat affected by the different features of the objects Zelinsky and Bisley, 2015). Therefore, a priority map is a representation of the visual world where a combination of bottom-up and topdown influences occur in order to process a visual scene .…”

“…In fact, the most evolved conception of the saliency maps by Itti and colleagues (Itti and Koch, 2001) changes the conception of saliency maps to comprehend that, although much of the model emphasizes the interplay between attention and scene processing, both top-down attentional bias and training also influence visual perception. More recent works also adopt the name priority map to denote that both bottom-up and top-down mechanisms have a role in the visual attention process, especially in selective attention (Fecteau and Munoz, 2006;Serences and Yantis, 2006;Bisley, Ipata, Krishna, Gee and Goldberg, 2009; Ipata, Gee, Bisley, and Golberg, 2009;Zelinsky and Bisley, 2015). …”

“…Recent conceptions of the priority map model argue that priority maps need to have a topological representation of the space; thus, PPC, FEF and SC must have them in order to hold priority maps (Zelinsky and Bisley, 2015). It seems that, under simple visual task conditions (i.e., visual search pop-out conditions), neural responses of these areas are similar; however, as visual complexity increases, in terms of object identification or task demands, differences between PPC and FEF response have been observed (Zelinsky and Bisley, 2015). For instance, LIP neurons are sensitive to set size changes, but his effect is not visible in FEF neurons (Balan, Oristaglio, Schneider and Gottlieb, 2008; Churchland, Kiani and Shandlen 2008;Mirpour and Bisley, 2012).…”