Scaling suprathreshold stimulus in chromatic threshold units: is i tan appropriate metric?

COLOMBO ELISA M.; O´DONELL, BEATRIZ M.

Konsberg

Congreso; The 21st Symposium of the INTERNATIONAL COLOUR VISION SOCIETY; 2011

Buskerud University College Konsberg Norway

Previous works (McKeefry et al, IOVS, 44: 2267-76, 2003; Zele et al, Vis. Res. 47:608-611, 2007; Vassilev et al, Vis.Res., 49: 524-29, 2009) still discuss the best way to scale a stimulus to provide an indication of its perceptual strength. Scaling suprathreshold stimuli in multiples of threshold has been used for the purpose of comparing suprathreshold performance for stimuli modulated along different dimensions. However, Zele y col,  suggest that expressing suprathreshold data in threshold units can produce outcomes artificially biased in favour of the mechanism with poorer sensitivity. While McKeefry et al showed that, if the stimulus is modulated on cardinal axes, the differences in reaction times are reduced when scaling is based on detection threshold. An alternative scaling metric is contrast, where stimulus is characterized by their physical properties, like Weber or Michelson contrast. Vassilev et al  said that the Weber contrast offers a more appropriate metric to specify suprathreshold stimuli. In order to add some piece of evidence to this discussion we modified the reaction time model proposed by O’Donell et al (Vis.Neurosc. 27: 119-129, 2010) replacing the Weber chromatic contrast -RMSi – by a multiple of chromatic threshold (RMSi/RMSit). The results seem to be contradictory since from this new analysis it turns out that color with chromatic direction equal to 90 deg appears as the most informative. Moreover, this model shows an inconsistent trend of the shared contrast gain which grows as the size decreases. Scaling suprathreshold stimuli in multiples of each observer’s color detection threshold seems not to be appropriate, favouring the chromatic direction with lower sensitivity and smallest size of the stimuli. Threshold sensitivity and reaction time are related to different features of the internal response generated by the visual stimulus. So, the Weber contrast offers a more appropriate metric for specifying suprathreshold stimuli.