Unconscious inhibition and facilitation at the objective detection threshold: Replicable and qualitatively different unconscious perceptual effects
Section snippets
Methodological and theoretical issues in demonstrating unconscious perception
Attempts to demonstrate unconscious perceptual influences typically rely on some version of the dissociation paradigm (Erdelyi and Erdelyi, 1985, Erdelyi, 1986). Usually, performance on two tasks is compared; one intended to index conscious perception, the other unconscious perception. In its classic and most common form, the dissociation paradigm seeks to obtain effects on the unconscious perception index despite null sensitivity on the conscious perception index, thus justifying inferences
But how should awareness be assessed?
Everyone agrees that conscious perception is positively related to stimulus intensity. Strong stimuli are clearly visible; as stimulus intensity is reduced (by varying stimulus duration and/or the intensity of masking), visual percepts become fainter and less distinct. Finally, when stimuli are sufficiently weak, conscious perception seems eliminated. But how, exactly, should one define the threshold for consciousness? It turns out that there are two basic alternatives.
The theoretical tension between subjective and objective threshold approaches
The very idea of objective threshold effects raises important theoretical and methodological questions. Namely, if subjective threshold approaches are valid, the robustly above-chance performance on direct tasks obtained under such conditions is due to unconscious influences. If so, reducing stimulus intensity to the point that direct task performance is at chance, as objective threshold methods require, should seriously reduce or even eliminate not only conscious but unconscious perceptual
Direct measures, unconscious perceptual influences, and the current paradigm
In the current paradigm, SDT detection is the conscious perception index, and forced-choice identification—also a direct task—is the unconscious perception index. Although using direct tasks to index unconscious influences is common (indeed, canonical) in subjective threshold approaches, objective threshold approaches typically use indirect tasks for this purpose. Indeed, it may seem particularly paradoxical to seek unconscious influences on identification given that it is often used to index
The original experiments
In the original experiments (Snodgrass et al., 1993a, Experiments 1 and 2; replicated by Van Selst & Merikle, 1993, Experiments 1 and 2), Preference×Strategy interactions were repeatedly obtained; there were no main effects. The pooled data are presented in Table 1. Simple effects analyses indicated that the interaction was primarily carried by a Preference congruity effect in the pop strategy: look preference participants (lookers) repeatedly performed below chance (the inhibition effect),
The current experiments—rationale and overview of organization and presentation
Although the original experiments were promising, we thought further replication of these effects advisable for two reasons: (1) objective threshold effects remain controversial (cf. Merikle and Reingold, 1998, Perruchet and Vinter, 2002), perhaps especially using direct tasks (cf. Holender, 1986); and (2) their exclusively bidirectional nature is highly unusual, perhaps reflecting important qualitative differences including a particularly powerful demonstration of uncontrollability, a widely
Participants
University students were paid $10–15, depending on the particular study. All participants had normal or corrected-to-normal vision and were native English speakers.
Apparatus and materials
Words were presented for 1 ms in a Gerbrand Model T3-8 Tachistoscope. Luminance was 10 fL for the stimulus field, fixation field, and ambient room light. Participants initially fixated a black dot on an otherwise blank stimulus card. The stimulus field (containing the word) was then flashed for 1 ms, followed immediately by the fixation
Experiments 1a and 1b
In Experiment 1a, we additionally collected event-related potential (ERP) brain wave measurements from participants while engaged in our paradigm; these ERP results are presented elsewhere (Shevrin, Snodgrass, Kushwaha, & Bernat, 2002). In Experiment 1b, we included female participants; our previous experiments used only males. For both experiments, we predicted that lookers would inhibit under pop instructions, and secondarily that poppers might facilitate under pop instructions—that is,
Discussion
Experiments 1a and 1b and the preliminary meta-analysis indicated that the Preference×Strategy interaction, including the looker inhibition effect, was quite reliable. With the benefit of this additional data, however, the form of this interaction was clarified importantly. Whereas it originally appeared that this interaction was carried by a pop strategy simple effect, inclusion of Experiments 1a and 1b suggest that it is best characterized as a looker simple effect, with little indication of
Methodological problems in unconscious perception research
Questions such as these return our attention to unresolved methodological issues in unconscious perception research—most importantly, ruling out alternative weak conscious perception explanations. There are two components to this difficulty—the exhaustiveness and null sensitivity problems (Reingold and Merikle, 1988, Reingold and Merikle, 1990. The exhaustiveness problem proper concerns whether the conscious perception index taps the relevant kind of conscious perception—that is, any and all
Inferring unconscious perception by falsifying the conscious-perception-only model
The core of the dissociation paradigm logic is that higher-level effects (e.g. semantic processing) should not be possible in the absence of lower-level effects (e.g. stimulus detection) if only conscious perception is involved. This is so because higher-level effects require more stimulus information (and stronger stimuli; see below) than lower-level effects; indeed, the former are built upon and presume the latter. For example, semantic analysis cannot occur unless the stimuli are at least
The fundamental qualitative difference
Given the above, skeptical concerns all make the same crucial prediction—that the conscious and unconscious perception indexes should correlate positively, or at least nonnegatively. More precisely, these alternative conscious-perception-only accounts predict that the size of the ostensibly unconscious effect (i.e. its absolute deviation from chance, whether positive or negative) should correlate positively, if at all, with the conscious perception index—and, in particular, that this should
Experiment 2
The primary change in Experiment 2 was the addition of an SDT detection task following the unconscious perception index (identification) trials. By having the same participants provide both detection and identification information, we could test the crucial skeptical prediction just described. If a positive correlation is obtained, alternative conscious perception accounts become much more plausible. Conversely, finding a negative relationship would contradict such skeptical accounts,
Discussion
Here, we propose an explanation for the Preference×Strategy interaction; more general implications are addressed in the General discussion below. As perusal of Table 4 indicates, the overall pattern suggests a Preference/Strategy congruity effect. Given this pattern, the obtained effects may reflect an unconscious attribution process. To begin with, the stimulus presentation on each trial presumably increases that word's activation relative to the other response alternatives. However, whether
General discussion
In our view, the current, clearly reliable findings exhibit qualitative differences, which have two major implications. First, they provide strong evidence against alternative conscious perception accounts. Second, they also suggest that objective and subjective threshold phenomena are likely fundamentally distinct, rather than the former simply being a weaker version of the latter.
But what does the negative relationship mean?
Above and beyond its salutatory methodological import, however, one might wonder about the substantive meaning of the negative relationship. To begin with, obtaining any relationship at all between the Preference×Strategy interaction and detection indicates that systematic variance is present in the latter as well as the former, despite the obtained null detection sensitivity. With this in mind, such variation could be due to conscious or unconscious perceptual influences on detection.
Unconscious perception and conscious control
Perhaps the most widely held qualitative difference hypothesis regarding conscious versus unconscious perception is that responses to conscious, but not unconscious, perceptions can be consciously controlled. In our view, there are good reasons to believe that subjective threshold effects are at least potentially controllable, whereas the current objective threshold effects are not—yielding another reason to believe that the two paradigms index fundamentally distinct constructs.
Objective versus subjective threshold approaches index qualitatively distinct processes
Given the foregoing, three independent qualitative differences convergently suggest that objective and subjective threshold approaches likely index fundamentally distinct processes. First, under objective threshold conditions, direct tasks exhibit no unidirectional component, but rather only bidirectional effects; in contrast, subjective threshold effects invariably include prominent unidirectional influences on direct tasks.8
Relevance to the tension between objective and subjective threshold approaches
The theoretical tension between objective and subjective threshold approaches, discussed above, derives from the exclusiveness problem, which seems to imply that all unconscious influences should decrease to zero as objective thresholds are approached. Perhaps, however, this applies only to subjective threshold effects; in contrast, objective threshold effects become larger, not smaller, as the ODT is approached. If objective and subjective threshold phenomena indeed qualitatively differ in
Objective thresholds, subjective thresholds, and phenomenal awareness
If the two methods indeed index distinct processes, what kind(s) of awareness do they index? To begin with, because objective threshold methods are more stringent, almost everyone would view such effects (if genuine) as phenomenally unconscious. In view of the above qualitative differences, however, one could question whether subjective threshold effects are phenomenally unconscious. Indeed, with normal populations, the only clearly established qualitative differences between subjective
Acknowledgements
This research was undertaken as part of a research program on conscious and unconscious processes in the Ormond and Hazel Hunt Memorial Laboratory, directed by Howard Shevrin and supported in part by gifts from Robert Berry. Special thanks to Mark Van Selst, Philip Merikle, and Steven Haase for making their raw data available. We also thank John Thompson for preparing the artwork.
References (84)
- et al.
Subliminal visual oddball stimuli evoke a P300 component
Journal of Clinical Neurophysiology
(2001) Paradox and cross purposes in recent work on consciousness
Cognition
(2001)- et al.
Task-induced strategies and near-threshold priming: Conscious influences on unconscious perception
Journal of Memory and Language
(1989) A process dissociation framework: Separating automatic from intentional uses of memory
Journal of Memory and Language
(1991)- et al.
Introspective physicalism as an approach to the science of consciousness
Cognition
(2001) - et al.
Effects of temporal cuing on residual visual discrimination in blindsight
Neuropsychologia
(1999) Conscious and unconscious perception: Experiments in visual masking and word recognition
Cognitive Psychology
(1983)Conscious and unconscious perception: An approach to the relations between phenomenal experience and perceptual processes
Cognitive Psychology
(1983)- et al.
Parallels between perception without attention and perception without awareness
Consciousness and Cognition
(1997) - et al.
Measuring the relative magnitude of unconscious influences
Consciousness and Cognition
(1995)
Perception without awareness: Perspectives from cognitive psychology
Cognition
Re-representing consciousness: Dissociations between experience and meta-consciousness
Trends in Cognitive Sciences
The dissociation paradigm and its discontents: How can unconscious perception or memory be inferred?
Consciousness and Cognition
The mediation of intentional judgments by unconscious perceptions: The influences of task strategy, task preference, word meaning, and motivation
Consciousness and Cognition
Absolute inhibition is incompatible with conscious perception
Consciousness and Cognition
Perception below the objective threshold?
Consciousness and Cognition
Conscious and unconscious processes: The effects of motivation
Consciousness and Cognition
Laboratory studies of behavior without awareness
Psychological Bulletin
Parts outweigh the whole (word) in unconscious analysis of meaning
Psychological Science
On a confusion about a function of consciousness
Behavioral and Brain Sciences
Priming with and without awareness
Perception & Psychophysics
Distinguishing conscious from unconscious perceptual processes
Canadian Journal of Psychology
Inhibitory processes in perceptual recognition: Evidence for a center-surround attentional mechanism
Unconscious perception: Attention, awareness, and control
Journal of Experimental Psychology: Learning, Memory, and Cognition
Preconscious processing
The response window method: Some problematic assumptions. Comment on Draine and Greenwald (1998)
Journal of Experimental Psychology: General
Replicable unconscious semantic priming
Journal of Experimental Psychology: General
Links between conscious awareness and response inhibition: Evidence from masked priming
Psychonomic Bulletin & Review
Psychoanalysis: Freud's cognitive psychology
Experimental indeterminancies in the dissociation paradigm of subliminal perception
Behavioral and Brain Sciences
Detection and recognition
Psychological Review
Signal detection theory and psychophysics
New Look 3: Unconscious cognition reclaimed
American Psychologist
Three cognitive markers of unconscious semantic activation
Science
Activation by marginally perceptible (‘subliminal’) stimuli: Dissociation of conscious from unconscious cognition
Journal of Experimental Psychology: General
Long-term memory versus contextual memory in unconscious number processing
Journal of Experimental Psychology: Learning, Memory, and Cognition
Confidence in word detection predicts word identification: Implications for an unconscious perception paradigm
American Journal of Psychology
A signal detection theory analysis of an unconscious perception effect
Perception & Psychophysics
Statistical methods for meta-analysis
Semantic activation without conscious identification in dichotic listening, parafoveal vision, and visual masking: A survey and appraisal
Behavioral and Brain Sciences
Bidirectional experimental effects
Psychological Methods
Cited by (72)
Larger capacity for unconscious versus conscious episodic memory
2021, Current BiologyUnconscious influence over executive control: Absence of conflict detection and adaptation
2018, Consciousness and CognitionPerception: Implicit, subliminal
2016, The Curated Reference Collection in Neuroscience and Biobehavioral PsychologyUnconscious biases in task choices depend on conscious expectations
2015, Consciousness and CognitionCitation Excerpt :Moreover, the non-conscious bias was much stronger at the SU threshold than at the OU threshold. These two facts suggest that both thresholds are “indexing different phenomena” (Snodgrass & Shevrin, 2006). A conservative interpretation of this difference would claim that SU is essentially conscious, since subjective reports could be misleading (Block, 2007; Hannula, Simons, & Cohen, 2005; Holender & Duscherer, 2004).