The impact of warning cues on target-discriminating decisions in continuous monitoring situations

In many real-world situations, sensory events of interest might appear at any moment requiring us to choose an appropriate action. This is especially challenging when stimuli are ambiguous or noisy, but a temporal cues can facilitated this process. The neural mechanisms underlying such temporally-uncertain, noise-embedded target discrimination and how warning cues dynamically adjust those mechanisms remains however unclear. We recently developed a neurally-informed (EEG), continuous bounded accumulation model, which, through adjustments to sensory criteria set on the evidence to be accumulated, was able to explain how subjects balance hits against inter-target false alarms in a pure detection situation (Geuzebroek et al 2023, eLife). Here, we extend this joint behavioural-neural analysis and modelling framework to discrimination decisions and use it to determine the computational adjustments in the decision-making process through which warning cues impact performance.Participants continuously monitored a moving-dot stimulus for 4-minute blocks, indicating the direction (left/right) of brief periods of coherent motion. Half of these targets were preceded by an auditory warning signal one second in advance. Behavioural modelling showed that participants respond to the cue by quickly lowering their decision bound, which is supported by a post-cue step-down in electroencephalographic signatures of motor preparation towards an action-triggering bound. However, more adjustments are found as signatures of expectation and attentional modulation are also strongly modulated. In ongoing work, additional neural signals such as the centro-parietal positivity (CPP) reflecting evidence accumulation and neurally-informed model comparison will allow us to map the global modulation of the cue on the decision-making processes.