slides (part 2) - Gestalt ReVision

Report
PERCEPTUAL
BISTABILITY AS A TOOL
FOR STUDYING UNCONSCIOUS VISUAL
PROCESSING
PIETER MOORS
Perceptual bistability
• Despite constant retinal input, perceptual
experience changes continuously
• The processing of the suppressed
stimulus can give insight into the neural
correlates of visual awareness.
Paradigms
• Motion-induced blindness
Paradigms
• Motion-induced blindness
• Binocular rivalry and continuous flash
suppression
– See also the session on binocular rivalry
Paradigms
• Motion-induced blindness
• Binocular rivalry and continuous flash
suppression
– See also the session on binocular rivalry
• (Visual masking, inattentional blindness,
change blindness, visual crowding,
attentional blink, object substitution
masking, …)
Measures
• Adaptation to the suppressed stimulus
• Priming by the suppressed stimulus
• Dominance/suppression durations
• Breaking suppression (CFS)
Overview
• Discuss findings on “low-level”, “midlevel”, and “high-level” aspects of
suppressed stimuli
• End with discussion on the convergence
and divergence between these findings
“Low-level” features
• The representation of very simple
features seems to be preserved during
suppression
– Orientation
(Blake & Fox, 1974; Blake et al., 2005; Montaser-Kouhsari et al.,
2004; Moradi & Koch, 2005, …)
– Spatial frequency
– Color
(Yang & Blake, 2012, …)
(Hofstoetter et al., 2004, Hong & Blake, 2009)
– Motion
(Lehmkuhle & Fox, 1975; O’Shea & Crassini, 1981, …)
“Mid-level” features
• Gestalt grouping cues can influence
dominance durations in
– MIB (Bonneh et al., 2001; Mitroff & Scholl,
2005)
Bonneh et al. (2001)
Mitroff & Scholl (2005)
“Mid-level” features
• Gestalt grouping cues can influence
dominance durations in
– MIB (Bonneh et al., 2001; Mitroff & Scholl,
2005)
– Binocular rivalry (Alais & Blake, 1999)
• Common fate
• Parallelism
• Good continuation
“Mid-level” features
• Mixed evidence for more complex
Gestalts
– E.g., Kanizsa stimulus
• Yes: Masking (Poscoliero et al., 2013); CFS (Wang et
al., 2013)
• No: BR (Sobel & Blake, 2003), CFS (Harris et al.,
2011)
Poscolieri et al. (2013)
Wang et al. (2012)
Sobel & Blake (2003)
Harris et al. (2011)
Moors et al. (in prep.)
• Divergence between direct and indirect
measures in rivalry and CFS.
• Collinear contours of the pacmen could
still drive the effect
Moors et al. (in prep.)
• Collinear contours of the pacmen could
still drive the effect
• Extracting the surface is critical for the
illusion
• Goal: Replicate effect from Wang et al.
(2012) and use control stimuli for surface
and collinear contours
• Stimulus set
Kanizsa
Regular
Irregular
Surface
Cross
• b-CFS experiment
Non dominant eye
Dominant eye
• Results
N = 20
Main effect Stimulus and Regularity (BF = 48)
“High-level” features
• Faces
– Adaptation in CFS
• Face shape only (Stein & Sterzer, 2011)
• Not identity (Moradi & Koch, 2005)
• Not gender or race (Amihai et al., 2011)
– b-CFS findings
• Inversion effects (Jiang et al., 2007)
• Familiarity effects (Gobbini et al., 2013)
“High-level” features
• Emotional faces
– Adaptation to emotional expressions (Adams
et al., 2010)
– Fearful faces break suppression faster (Yang
et al., 2007)
• Potential spatial frequency confound? (Stein &
Sterzer, 2011; Gray et al., 2013)
“High-level” features
• Semantic processing
– In CFS
• Scene congruency biases suppression (Mudrik et al.,
2010, 2011)
• For word stimuli
– Prime – target congruency (Costello et al., 2009)
– Emotional vs. neutral words (Yang & Yeh, 2011; Sklar
et al., 2012)
“High-level” features
• Semantic processing
– No study addressed whether words per se
break suppression faster than pseudo/nonwords
– If words are processed, a word frequency
effect is expected to be observed
Heyman and Moors (2014)
• Experiment 1
– Word type (word vs. pseudo-word)
– Word frequency
• Experiment 2
– Pseudo words still readable
• Words vs. non-words
– Letter familiarity
• Upright vs. inverted
Heyman and Moors (2014)
Null model (BF = 26)
N = 18
Null model (BF = 11)
Heyman and Moors (2014)
Null model (BF = 15)
N = 20
Null model (BF = 10)
Discussion
• Unconscious visual processing?
– Findings often depend on the paradigm
– Paradigms differ in their suppression
mechanisms
• What does this imply for “unconscious”
visual processing in general?
Discussion
• Are we learning something about
unconscious visual processing?
• Or are we learning something about the
level at which a suppression method acts
and the representation of the stimulus at
that level?
• Example: Kanizsa stimulus
– CFS vs. masking vs. MIB
Discussion
• This point is increasingly being recognized
– Breitmeyer (2014)
• “Much of brain imaging research relies on the use of
one or another of the many noninvasive ways of
rendering stimuli invisible. Without knowing at which
cortical level of processing these blinding techniques
exert their effects, one cannot make unequivocal
claims as to what are or are not NCoCs and NCoUs.”
Discussion
Discussion
– Fogelson et al. (2014)
• Continuous flash suppression versus Chromatic
flicker fusion:

similar documents