Feature Integration: An Emperical Study


Introduction to Cognitive Sciences

Prof. Amitabha Mukarjee




Abstract

Features are integrated before the objects are made sense of. Taking for granted that features are integrated, I performed test as to what is the accuracy of subjects in recognizing random numbers generated in para-foveal cone, in both horizontal and vertical direction. I also checked users accuracy at recognizing the random numbers by varying the contrast ratio of the numbers with respect to the background. 
The results are pointing to the fact that we are better ate recognizing the objects in horizontal direction than in the vertical direction. Also contrast doesn’t really affect the accuracy of recognition till about a certain threshold contrast given that the random numbers were flashed for a limited period of time.

Summary

The results of the study point to differential horizontal, vertical regions of visual acuity and expected decrease in accuracy as the contrast went down.

Acknowledgements

I would like to acknowledge Prof. Amitabha Mukerjee for teaching a higly informative and instructive course “Cognitive Sciences”. I would also like to thank M Seetha Ramaiah for taking care of all the doubts and problems that we faced. I would also like to thank the subjects of my study, Vijay Kumar, K. Prashant, Rahul Kataria, Suraj Bhamare, Abhishek Ranjan, Deepak Singh for being patient subjects in the study. It took more than one hour on each subject, and I express my gratitude to them for having beared with me of watching a blind fixation point on the screen for 1 hour without any incentives. I would also like to thank Vijay Kumar again for being my first subject and helping evolve the experiment.
Next I would like to thanks Sashank Pisupati for valuable inputs in discussion at the time of poster presentation.
I would thank IIT Kanpur for facilities, and last but not the least my family and friends for making me what I am.

Important Links

Project Proposal
Presentation Slides(pdf)
Presentation Slides(pptx)
Poster
Project Report
Project Flash Programs
Video of a subject in experiment

 




References

  1. Freeman, J., & Pelli, D. G. (2007). An escape from crowding. Journal of Vision 7(2):22, 1-14.
  2. Hagenaar, R., & Van der Heijden, A. (1986). Target-noise separation in visual selective attention. Acta Psychologica, 62, 161-176.
  3. Pelli, D. G., Melanie, P., & Najib, J. M. (2004). Crowding is unlike ordinary masking: Distinguishing feature integration from detection. Journal of Vision 4, 1136-1169.
  4. Poder, E. (2006). Crowding, feature integration, and two kinds of ‘‘attention’’. Journal of Vision, 6, 163-169.
  5. Strasburger, H., Harvey Jr, L. O., & Rentschler, I. (1991). Contrast thresholds for identification of numeric characters in direct and eccentric view. Perception & Psychophysics 49 (6), 495-508.
  6. Treisman, A. M., & Gelade, G. (1980). A Feature Integration Theory of Attention. Cognitive Psychology 12, 97-136.