Skip to main content
Log in

The eyes do not have it after all? Attention is not automatically biased towards faces and eyes

  • Original Article
  • Published:
Psychological Research Aims and scope Submit manuscript

Abstract

It is commonly accepted that attention is spontaneously biased towards faces and eyes. However, the role of stimulus features and task settings in this finding has not yet been systematically investigated. Here, we tested if faces and facial features bias attention spontaneously when stimulus factors, task properties, response conditions, and eye movements are controlled. In three experiments, participants viewed face, house, and control scrambled face–house images in an upright and inverted orientation. The task was to discriminate a target that appeared with equal probability at the previous location of the face, house, or the control image. In all experiments, our data indicated no spontaneous biasing of attention for targets occurring at the previous location of the face. Experiment 3, which measured oculomotor biasing, suggested a reliable but infrequent saccadic bias towards the eye region of upright faces. Importantly, these results did not reflect our specific laboratory settings, as in Experiment 4, we present a full replication of a classic finding in the literature demonstrating reliable social attention bias. Together, these data suggest that attentional biasing for social information is task and context mediated, and less robust than originally thought.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Fig. 1
Fig. 2
Fig. 3
Fig. 4
Fig. 5
Fig. 6
Fig. 7
Fig. 8

Similar content being viewed by others

Notes

  1. We thank Markus Bindemann for providing us with the original stimuli.

  2. Twenty-eight additional naïve participants were asked to rate images of various faces and houses using a Likert scale ranging from 1- Very Unattractive to 6- Very Attractive. The Face and House images that were used here received equivalent attractiveness ratings (Face M = 2.93, SD = 0.77; House M = 2.96, SD = 0.96), which did not differ statistically, t(27) = 0.17, p = 0.87, dz = 0.03.

  3. Confirming no speed-accuracy tradeoffs, an additional ANOVA examining mean accuracy rates with the same factors confirmed higher overall accuracy for short relative to long cue-target intervals [Cue-target interval, F(3,87)=9.23, p < 0.001, \(\eta _{{\text{p}}}^{2}\) = 0.24; 250 ms vs. 560 ms and 1000 ms, ts > 3.36, ps < 0.01, dzs > 0.61; 360 ms vs. 1000 ms, t(29)=2.78, p = 0.036, dz = 0.51; all other ps > 0.07, dzs < 0.44] and overall lowest accuracy for targets appearing at the location of the neutral cues [Target position, F(5,145)=29.74, p < 0.001, \(\eta _{{\text{p}}}^{2}\) = 0.51; upper and lower neutral vs. all, ts > 5.48, ps < 0.001, dzs > 1.00; all other ps > 0.56, dzs < 0.33]. A significant interaction between Cue orientation and Face position, F(1,29)=4.46, p = 0.043, \(\eta _{{\text{p}}}^{2}\) = 0.13, indicated lower overall accuracy when inverted faces were presented in the right visual field, t(29)=3.29, p = 0.006, dz = 0.60; other p = 0.76, dz = 0.06. No other effects involving Face position and Target position were significant, Fs < 2.80, ps > 0.11, \(\eta _{{\text{p}}}^{2}\) < 0.08.

  4. Analyses of response accuracy once again indicated no speed-accuracy trade-offs. The ANOVA returned a marginal main effect of Cue-target interval, F(3,87)=2.67, p = 0.052, \(\eta _{{\text{p}}}^{2}\) = 0.08, with higher accuracy for targets appearing at short relative to long cue-target intervals [250 ms vs. 1000 ms, t(29)=2.83, p = 0.048, dz = 0.52; all other ps > 0.34, dzs < 0.35]. A main effect of Target position, F(5,145)=45.18, p < 0.001, \(\eta _{{\text{p}}}^{2}\) = 0.61, once again indicated lower accuracy for targets at the previous location of both neutral cues [upper and lower neutral vs. all, ts > 5.90, ps < 0.001, dzs > 1.08]. Lower accuracy was also found for targets occurring at the previous location of the mouth vs. eye cues, t(29) = 3.11, p = 0.028, dz = 0.57; all other ps > 0.10, dzs < 0.46. No other effects or interactions were reliable, all Fs < 1.63, ps > 0.18, \(\eta _{{\text{p}}}^{2}\) < 0.05.

  5. No speed-accuracy trade-off was evident. The same ANOVA conducted on accuracy revealed a main effect of Target position, F(5,145)=15.74, p < 0.001, \(\eta _{{\text{p}}}^{2}\) = 0.35, with lower accuracy for targets appearing in the previous location of the upper and lower neutral cues vs. all others [ts > 3.60, ps < 0.008, dzs > 0.66; all other ps > 0.99, dzs < 0.25]. An interaction between Cue orientation and Target position, F(5,145) = 3.42, p = 0.006, \(\eta _{{\text{p}}}^{2}\) = 0.11, indicated lower accuracy for targets that occurred at the previous location of the neutral cues (upper, lower) vs. the eyes, mouth, and house top for upright cues [ts > 3.60, ps < 0.011, dzs > 0.66; all other ps > 0.24, dzs < 0.43] and lower accuracy for targets that occurred at the previous location of both neutral cues (upper, lower) vs. the eyes, mouth, and house bottom for inverted cues [ts > 3.57, ps < 0.011, dzs > 0.65; all other ps > 0.30, dzs < 0.40]. No other main effects or interactions were found, all other Fs < 2.64, ps > 0.12, \(\eta _{{\text{p}}}^{2}\) < 0.08.

References

Download references

Acknowledgements

Many thanks to J. Michelin and E. Bossard for their help with this project. The data sets from the current study are available from the corresponding author on reasonable request.

Funding

This study was funded by Natural Sciences and Engineering Research Council of Canada (NSERC) and NSERC-CREATE graduate fellowships to EJP, NSERC, and Social Sciences and Humanities Research Council (SSHRC) grants to EB and JR, and W. Dawson award to JR.

Author information

Authors and Affiliations

Authors

Corresponding author

Correspondence to Effie J. Pereira.

Ethics declarations

Conflict of interest

The authors declare that they have no conflicts of interest.

Ethical approval

All procedures performed in these studies involved human participants and were in accordance with the ethical standards of the institutional and/or national research committee, and with the 1964 Helsinki declaration and its later amendments or comparable ethical standards.

Informed consent

Informed consent was obtained from all individual participants included in the study.

Additional information

Publisher’s Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Rights and permissions

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Pereira, E.J., Birmingham, E. & Ristic, J. The eyes do not have it after all? Attention is not automatically biased towards faces and eyes. Psychological Research 84, 1407–1423 (2020). https://doi.org/10.1007/s00426-018-1130-4

Download citation

  • Received:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s00426-018-1130-4

Navigation