Is there semantic conflict in the Stroop task? Further evidence from a modified two-to-one Stroop paradigm combined with singleletter coloring and cueing

International audienceThis research addressed current controversies concerning the contribution of semantic conflict to the Stroop interference effect and its reduction by single-letter coloring and cueingprocedure (SLCC). On the first issue it provides, for the first time, unambiguous evidence for a contribution of semantic conflict to the (overall) Stroop interference effect. The reported data remained inconclusive on the second issue, despite being collected in a considerable sample and analyzed with both classical (frequentist) and Bayesian inferential approaches. Given that in all past Stroop studies, semantic conflict was possibly confounded with either response conflict (e.g., when semantic-associative items [SKY blue ] are used to induce semantic conflict) or with facilitation (when color-congruent items [BLUE blue ] are used as baseline to derive a magnitude for semantic conflict), its genuine contribution to the Stroop interference effect is the most critical result reported in the present study. Indeed, it leaves no doubt-in complete contrast to dominant single-stage response competition models (e.g., Roelofs, 2003)-that selection occurs at the semantic level in the Stroop task. The immediate implications for the composite (as opposed to unitary) nature of the Stroop interference effect and other still unresolved issues in the Stroop literature are outlined further