Repository logo

Infoscience

  • English
  • French
Log In
Logo EPFL, École polytechnique fédérale de Lausanne

Infoscience

  • English
  • French
Log In
  1. Home
  2. Academic and Research Output
  3. Conferences, Workshops, Symposiums, and Seminars
  4. Serial dependence across features and objects
 
conference poster not in proceedings

Serial dependence across features and objects

Ceylan, Gizay  
•
Herzog, Michael  
•
Pascucci, David  
July 11, 2020
12th Forum of Neuroscience (FENS 2020)

Aims Visual perception is systematically biased toward stimuli seen in the recent past, a phenomenon known as serial dependence (SD). It is widely believed that SD reflects a continuity field in vision to promote visual stability over time. A main question, however, is whether SD truly affects perception per se or rather reflects a bias at higher-level processing stages. In this work, we investigated whether SD occurs even between different visual features and objects. The specificity of SD to basic visual features would support an early perceptual effect; strong SD between even different objects would instead support post-perceptual biases. Methods In two behavioral experiments, human participants performed an orientation adjustment task in which stimuli could change in a single feature (e.g., spatial frequency) or at the level of the entire object (e.g., Gabor patch or symmetric dot patterns). To measure SD, we quantified the deviation of adjustment errors toward previous orientations during trials with either the same or a different visual feature/object. Results We found systematic biases in adjustment errors toward previous stimuli that are independent of changes in features and objects. The effect size of this bias was exclusively modulated by the strength of the present sensory signals and was maximum for low spatial frequency stimuli. Conclusions Our results demonstrate that SD is located beyond early perceptual processing, since it occurs even for different features and highly dissimilar stimuli. We conclude that SD emerges at high-level processing stages where task-relevant representations are dissociated from the physical attributes of stimuli.

  • Files
  • Details
  • Metrics
Loading...
Thumbnail Image
Name

Ceylan_et_al_FENS2020_Poster.pdf

Type

N/a

Access type

openaccess

License Condition

CC BY-NC-ND

Size

1.7 MB

Format

Adobe PDF

Checksum (MD5)

2166eb6d8ffa3c715c1ed4281b56d54f

Logo EPFL, École polytechnique fédérale de Lausanne
  • Contact
  • infoscience@epfl.ch

  • Follow us on Facebook
  • Follow us on Instagram
  • Follow us on LinkedIn
  • Follow us on X
  • Follow us on Youtube
AccessibilityLegal noticePrivacy policyCookie settingsEnd User AgreementGet helpFeedback

Infoscience is a service managed and provided by the Library and IT Services of EPFL. © EPFL, tous droits réservés