A potential for distraction: Using task-irrelevant complex environment sounds to probe closed-loop control demands

Distractions are often viewed as a negative occurrence in the study of human factors. Nonetheless, the ability to disengage and redirect our attention to an unexpected event is vital to our survival. It allows us to detect possible threats or more rewarding goals. In my talk, I will speak on the event-related potential (ERP; distraction potential) that is involuntarily generated by the appearance of a complex sound in our environment, even when the sound bears no relevance to the task at hand and does not require an explicit response. I will expand on the characteristic waveform of the distraction potential and the factors that influence it. More importantly, I will propose how the distraction potential could be exploited towards understanding the task demands, particularly of manipulating closed-loop control systems (e.g., vehicles). Environment sounds can represent events of interest even if it is not apparent how or why they are task-relevant. In the context of driving, this could be a crying child in the backseat or a non-descript engine squeak. Even if a behavioral response is not required, the brain involuntarily responds, giving rise to a characteristic ERP waveform that has been termed the distraction potential (Escera & Corral, 2003). Like the ERPs produced by task-relevant pure tones, namely those that participants are explicitly instructed to monitor and discriminate for, distraction potentials of task-irrelevant environment sounds depict an early negative deflection (cf., N1), followed by a late positive deflection (cf., P300). Unlike a typical auditory ERP, the late positive deflection of a distraction potential typically consists of two distinct positive deflections where a single P300 would typically be. The two P3 components (namely the early and late novelty P3 components) of a distraction potential are sensitive to the demands of an ongoing task, even though the environment sounds that generate them do not require a response. Indeed, observers are explicitly instructed to ignore these sounds. We have shown that visuo-motor steering demands attenuate the amplitudes of the early and late novelty P3 components of this distraction potential (Scheer, Bülthoff, Chuang, 2016). In contrast, the P300s generated by pure tones are sensitive to steering demands only if participants are instructed to discriminate target tones from distractor tones within a dual-task paradigm (Wickens, Kramer, Vanasse, & Donchin, 1983). This means that task-irrelevant environment sounds can be used as probes for evaluating the demands of a visuo-motor steering task, without necessitating a secondary auditory discrimination task that could interact with and modify the demands of steering itself. Nonetheless, the question remains: Which of the two P3 components of a distraction potential are likely to be comparable to the well-established P300 generated to task-relevant pure tones? Two reasons lead us to believe that the late novelty P3 generated by task-irrelevant environment sounds is likely to resemble the P300 generated by task-relevant pure tones. First, we recently reported that instructing participants to pay attention to the auditory channel selectively increased the late novelty P3, but not the early novelty P3 component, of the distraction potential (Scheer, Bülthoff, Chuang, 2018). Thus, the late novelty P3 reflects the strategic allocation of attentional resources to a relevant modality regardless of the stimulus’ task relevance. Second, varying the complexity of the visuo-motor task’s control dynamics selectively attenuates only the late novelty P3 component of the distraction potential, whilst having no effect on the early novelty P3 (Scheer & Chuang, under review). Other manipulations of steering difficulty, such as control disturbances, do not have a similar impact. These findings mirror those found in dual-task paradigms whereby similar conflicts were found on the P300 potentials generated by infrequent task-relevant tones (Wickens et al., 1983). Thus, we have proposed that the late novelty P3 component generated by complex and task-irrelevant sounds is comparable to the better established P300 potentials. Visuo-motor steering exerts a cross-modal decrement to auditory ERPs when it demands more executive working memory resources. This can be indexed either by the P300 of a task-relevant pure tone target or the late novelty P3 of a task-irrelevant environment sound. Operationally, this could give rise to the phenomenon of “inattentional deafness”, whereby highly trained pilots fail to recognize unexpected yet highly critical sounds (i.e., warning of landing gear failure; . Using the late novelty P3 component allows us to evaluate ongoing task demands without disrupting the task that is under investigation by introducing a secondary task. This offers researchers the opportunity of inobtrusively evaluating what is broadly referred to as the “cognitive workload” of more complex tasks, from playing Tetris (Miller, Rietschel, McDonald, & Hatfield, 2011) to flight control (Jacquess et al., 2017)