Abstract
In a previous study, we have shown that the sensitivity difference between luminance- (LM) and contrast-modulated (CM) stimuli results from a difference of internal equivalent noise and not from a difference of calculation efficiency. The objective of the present study was to seek the source of the internal noise limiting the sensitivity to CM stimuli. Three types of noise were used: band-pass LM noise near the carrier (LMN-carrier) or envelope (LMN-envelope) spatial frequency and band-pass CM noise near the envelope spatial frequency (CMN-envelope). For the five observers, the noise contrasts were adjusted to increase the detection thresholds by 0.5 log units for their respective stimuli: carrier, LM and CM. LM and CM detection thresholds were subsequently evaluated in these three noise conditions using a constant stimuli paradigm. As expected, LMN- and CMN-envelope increased the detection threshold of their respective stimulus. However, no cross-type interactions were found: LMN- and CMN-envelope had no significant impact on CM and LM stimuli detection respectively, and LMN-carrier did not affect the detection thresholds of LM and CM stimuli. This double dissociation is strong evidence suggesting that both stimuli are processed, at least partially, by separate mechanisms and that they are not merged after a second-order rectification applied to CM stimuli. The results also suggest that, in the tested conditions, pre-rectification noise affecting the carrier visibility is not a limiting factor for CM stimuli sensitivity.
This research was supported by NSERC graduate fellowship to RA and NSERC operating grant to JF