EBRU ECEM TAVACIOGLU¹ (), MATTHIAS GAMER¹, JEREMY WOLFE^2,3; ¹University of Würzburg, Germany, ²Brigham and Women's Hospital, ³Harvard Medical School

Human foraging in natural environments requires continuously allocating attention across space while weighing potential gains against potential costs. Although visual foraging paradigms have revealed much about how observers guide search across extended scenes, far less is known about how people decide where to forage, given spatially varying rewards and threats having a different spatial layout. Here, we examined how distributed value and aversive sound jointly shaped attentional allocation and harvesting behavior during continuous visual foraging. Participants completed a grid-based foraging paradigm in which reward magnitude (1–5 points) and auditory threat level (0–4 intensity) followed orthogonal gradients across a 20×20 display. In a baseline block, rewards were uniform and no sound was presented; in a gradient block, both reward and sound values were spatially structured. Participants freely harvested targets across multiple patches. Click-level behavior was extracted from all cell responses and aggregated into 5×5 spatial bins reflecting reward and sound levels. Relative to the baseline block, the gradient block led to higher click rate, more hits and false alarms (FA), reduced time per patch, and decreased positive predictive value (PPV) for each click. Critically, spatial heatmaps revealed that these effects were not uniformly distributed. Performance deteriorated most notably in the high-value regions, which showed a marked increase in FAs and a drop in PPV, whereas behavior in low-value/low-sound regions remained comparatively stable. These results visualize how distributed value and threat can jointly shape harvesting decisions, influencing both overall click behavior and the spatial distribution of hits and errors. Since behavioral adjustments were not uniform across the grid, we can reject the hypothesis that threat simply reduces overall performance. Together, these findings demonstrate that aversive sound impairs the efficient exploitation of high-value areas, reducing the efficiency of harvesting.