Learning rational temporal eye movement strategies

Abstract

During active behavior humans redirect their gaze several times every second within the visual environment. Where we look within static images is highly efficient, as quantified by computational models of human gaze shifts in visual search and face recognition tasks. However, when we shift gaze is mostly unknown despite its fundamental importance for survival in a dynamic world. It has been suggested that during naturalistic visuomotor behavior gaze deployment is coordinated with taskrelevant events, often predictive of future events, and studies in sportsmen suggest that timing of eye movements is learned. Here we establish that humans efficiently learn to adjust the timing of eye movements in response to environmental regularities when monitoring locations in the visual scene to detect probabilistically occurring events. To detect the events humans adopt strategies that can be understood through a computational model that includes perceptual and acting uncertainties, a minimal processing time, and, crucially, the intrinsic costs of gaze behavior. Thus, subjects traded off event detection rate with behavioral costs of carrying out eye movements. Remarkably, based on this rational bounded actor model the time course of learning the gaze strategies is fully explained by an optimal Bayesian learner with humans' characteristic uncertainty in time estimation, the wellknown scalar law of biological timing. Taken together, these findings establish that the human visual system is highly efficient in learning temporal regularities in the environment and that it can use these regularities to control the timing of eye movements to detect behaviorally relevant events.eye movements | computational modeling | decision making | learning | visual attention W e live in a dynamic and ever-changing environment. To avoid missing crucial events, we need to constantly use new sensory information to monitor our surroundings. However, the fraction of the visual environment that can be perceived at a given moment is limited by the placement of the eyes and the arrangement of the receptor cells within the eyes (1). Thus, continuously monitoring environmental locations, even when we know which regions in space contain relevant information, is unfeasible. Instead, we actively explore by targeting the visual apparatus toward regions of interest using proper movements of the eyes, head, and body (2-5). This constitutes a fundamental computational problem requiring humans to decide sequentially when to look where. Solving this problem arguably has been crucial to our survival from the early human hunters pursuing a herd of prey and avoiding predators to the modern human navigating a crowded sidewalk and crossing a busy road.So far, the emphasis of most studies on eye movements has been on spatial gaze selection. Its exquisite adaptability can be appreciated by considering the different factors influencing gaze, including low-level image features (6), scene gist (7) and scene semantics (8), task constraints (9), and extrinsic rewards (10); al...

Humans quickly learn to blink strategically in response to environmental task demands

“…where µ, σ 2 and p are the parameters of the mixture distribution generating the events, θ is a location during the lap, and n(θ) is the average number of events left at each location during a particular lap (see Supporting Information for details). We weighted very short IBIs using a cumulative Gaussian (32). This accounts for motor delays, making two blinks occurring with close to zero IBI very unlikely.…”

“…One reason might be that environmental regularities and task-related costs are usually complex and unknown. The lack of quantitative models is surprising, considering the strong contingencies between environmental statistics and gaze behavior, which have been explained successfully through modeling (29)(30)(31)(32). As with blink rates, few computational approaches exist that describe the temporal course of blinking.…”

Mapping the unknown: The spatially correlated multi-armed bandit

“…Local search. While simple, a tendency to stay local to the previous search decision-regardless of outcome-has been observed in many different contexts, such as semantic foraging (Hills, Jones, & Todd, 2012), causal learning (Bramley, Dayan, Griffiths, & Lagnado, 2017), and eye movements (Hoppe & Rothkopf, 2016). We use inverse Manhattan distance (IMD) to quantify locality:…”

Discriminative control of saccade latencies

“…This indicates that monkeys were able to discriminate among the various contingencies that in return came to control the saccades. Research also demonstrated that human observers may learn the temporal properties of a dynamical environment to allocate their gaze toward a specific region based on the associated frequency of reinforcement (Hoppe & Rothkopf, 2016). A similar result was obtained in a latency-contingent paradigm in which changes in reinforcement contingencies induced changes in saccade latency distributions (Vullings & Madelain, 2018).…”