Inhibition of return (IOR) meets stimulus-response (S-R) binding: Manually responding to central arrow targets is driven by S-R binding, not IOR

ABSTRACT

Localizing targets repeating or changing their position typically leads to a benefit for location changes, that is, inhibition of return (IOR). Yet, IOR is mostly absent when sequentially responding to arrows pointing to the left or right. Previous research suggested that responding to central arrow targets resembles a discrimination response. For the latter, action control theories expect the modulation of response repetitions and changes by task-irrelevant feature repetitions and changes (e.g., colour), caused by stimulus-response (S-R) binding – a modulation typically absent in localization performance. In the current study, participants gave left and right responses to peripheral targets repeating or changing their position, and to central arrow targets repeating or changing their pointing direction. Targets could repeat or change their colour. For central targets, responses were heavily modulated by colour repetitions and changes, suggesting S-R binding. No S-R binding, but only IOR was found for peripheral targets. Analysis of reaction time percentiles suggested that this pattern was not caused by fast response execution. These results show that S-R binding approaches allow to explain effects typically discussed in the context of attentional orienting, highlighting the similarities of two research strands working in parallel for years without much of exchange.

KEYWORDS:

• Attention
• inhibition of return
• stimulus-response binding
• action control

Acknowledgements

We would like to thank Matthew D. Hilchey, Iring Koch, and Juan Lupiáñez for helpful reviews that improved the manuscript. Research was partially funded by research grant SCHO 2000/1-1 awarded to Lars-Michael Schöpper by the Deutsche Forschungsgemeinschaft.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Data availability statement

Data of the experiment is available at http://dx.doi.org/10.23668/psycharchives.12362. Code for analysis of the experiment is available at http://dx.doi.org/10.23668/psycharchives.12361.

Notes

1 Note, however, that a “no-response tag” could also be integrated as some sort of response-equivalent into an event file (see Frings et al., 2020). In turn, no response to the cue followed by a response to the (potentially identical) target could be processed as a response change (see Mondor & Leboe, 2008; and Schöpper & Frings, 2022, for a discussion of cue-target vs. target-target designs when investigating detection performance).

2 Results remained stable by applying different cutoff-criteria (1.5 interquartile range), as well as using the medians without temporal cut-offs for calculations.

3 Using an ANOVA with Greenhouse-Geisser correction due to violated sphericity, the main effect of percentile reached significance, F(2.042, 38.791) = 3.23, p = .049, ${\eta }_p^2$ = .15. Note that this effect shows that the calculated binding effect increasingly gets negative with increasing percentile – thus being the opposite of what action control theories would predict but in accordance with non-spatial feature change benefits (see General Discussion). However, in the main text we report the MANOVA as it is unaffected by violations of sphericity.