Listening strategy determines how neural signals are filtered in a “spatial attention” task

Often, spatial attention filters out “distractors,” reducing cortical responses to sounds from an unattended direction. Here, in a nominal “spatial selective attention” task, listeners let through all potential targets before evaluating location. Participants heard a rapid, randomly ordered sequence comprising the words “bash,” “dash,” and “gash.” Individual words were spatialized left or right using either interaural level differences (ILDs) or interaural time differences (ITDs) derived from head-related transfer functions (±5° or ±15°). Listeners were asked to respond to “bash” in the target direction while we measured evoked responses using electroencephalography and hemodynamic responses over prefrontal cortex (PFC) using functional near-infrared spectroscopy. Rather than observing larger onset (P1-N1) responses evoked by words in the target direction, “bash” evoked larger onset responses than “dash” or “gash,” irrespective of direction. All “bashes” elicited a late “target recognition” (P3) response, but it was significantly larger for words from the target direction. PFC responses were larger with ±15° ITDs/ILDs than ±5°, unlike previous studies showing an inverse relationship between difficulty and PFC activity. Rather than relying on spatial selective attention, our listeners filtered words based on phonetic content, then evaluated the location of potential targets, highlighting the influence of task demands on listening strategy.