Introspection and subliminal perception

Subliminal perception (SP) is today considered a well-supported theory stating that perception can occur without conscious awareness and have a significant impact on later behaviour and thought. In this article, we first present and discuss different approaches to the study of SP. In doing this, we claim that most approaches are based on a dichotomic measure of awareness. Drawing upon recent advances and discussions in the study of introspection and phenomenological psychology, we argue for both the possibility and necessity of using an elaborated measure of subjective states. In the second part of the article, we present findings where these considerations are implemented in an empirical study. The results and implications are discussed in detail, both with reference to SP, and in relation to the more general problem of using elaborate introspective reports as data in relation to studies of cognition.     

Experimenter bias and subliminal perception.

It has been suggested that subliminal perception phenomena may be in part due to experimenter bias effects. Two studies that obtained positive evidence of subliminal perception were therefore replicated with experimenters tested under blind and not blind conditions. There was only marginal support for the subliminal perception hypothesis and, although there were fairly clear indications of diffuse experimenter effects, the evidence for the experimenter bias explanation of subliminal perception was not strong. The need for more extensive replications of subliminal perception researches is noted. It is argued that the experimenter bias hypothesis lacks detail and generality; it is essential for it to be examined in the context of theoretically substantial issues.     

Musical duplex perception: perception of figurally good chords with subliminal distinguishing tones.

In a variant of duplex perception with speech, phoneme perception is maintained when distinguishing components are presented below intensities required for separate detection, forming the basis for the claim that a phonetic module takes precedence over nonspeech processing. This finding is replicated with music chords (C major and minor) created by mixing a piano fifth with a sinusoidal distinguishing tone (E or E flat). Individual threshold intensities for detecting E or E flat in the context of the fixed piano tones are established. Chord discrimination thresholds defined by distinguishing tone intensity were determined. Experiment 2 verified masked detection thresholds and subliminal chord identification for experienced musicians. Accurate chord perception was maintained at distinguishing tone intensities nearly 20 dB below the threshold for separate detection. Speech and music findings are argued to demonstrate general perceptual principles.     

Changing the perception of behavioral properties by subliminal presentation

Profiles designed by Pittenger and Shaw were used in a tachistoscopic experiment. Subjects were asked to make estimates of age of a profile of a 10-yr.-old boy preceded subliminally by an older (old group) or younger (young group) profile or by a blank card (control group). Results show the mean estimate of the young group to be significantly smaller than that of the control group. The mean estimate of the old group was not significantly greater than that of the control group. The results are discussed in the framework of the debate on direct versus indirect perception. Marcel's 1983 data and the present results indicate that higher level perceptual processing (the perception of behavioral properties) can precede lower level perceptual processing (the perception of physical properties). By assuming that people do perceive directly the behavioral properties of an event, the direct approach avoids the danger of putting too much emphasis on intelligence in perception and at the same time avoids the homunculus trap.     

Threshold variability in subliminal perception experiments: fixed threshold estimates reduce power to detect subliminal effects.

In studies of subliminal perception, a threshold for awareness is usually estimated for each S (e.g., the stimulus duration that yields chance discrimination performance). This article shows that if the S's true threshold varies randomly from trial to trial, estimation of a fixed threshold has important consequences. Specifically, the estimated threshold approaches the minimum of the distribution of true thresholds, and the power of an experiment to detect subliminal perception is reduced. Because there is much evidence that true thresholds vary across trials, threshold variability may have seriously weakened previous experiments that tested for subliminal perception. Fortunately, it is possible to estimate the maximum power loss resulting from threshold variability and therefore to design an experiment with acceptable power even in the presence of variability.     

The subliminal perception of movement and the course of autokinesis

The course of autokinesis is shown to be sensitive to the real movement of a surrounding stimulus. With the supraliminal presentation of this stimulus, apparent movement in a direction opposite to that of the real movement is induced. With the subliminal presentation of the same stimulus the real movement serves to inhibit autokinesis by inducing brief periods of stationarity between the phases of upward and downward apparent movement. The results confirm previous findings that the movement of a stimulus may be discriminated without there being any perceptual (phenomenal) adjunct.     

Perceptual organization without perception. The subliminal learning of global contour

A critical step in visual perceptual processing is integrating local visual elements into contours so that shapes can be derived from them. It is often assumed that contour integration may reflect hardwired coding of low-level visual features. In this study, we present novel evidence indicating that integration of local elements into contours can be learned subliminally, despite being irrelevant to the training task and despite the local properties of the display varying randomly during training. Learning occurred only when contours were consistently paired with task-relevant targets--echoing the findings of previous studies on subliminal learning of low-level features. Our data indicate that task-irrelevant, exposure-based learning extends beyond local low-level visual features and may play a critical role at multiple levels of visual perceptual organization.     

The subliminal perception of movement and the 'suppression' in binocular rivalry.

An analogy is drawn between the perceptual limitation that characterizes the dichotic listening paradigm and the 'suppression' that occurs in binocular rivalry when different stimuli are presented to the two eyes. An experiment is reported which focuses on the fate of the information residing in a suppressed eye (unattended channel) during binocular rivalry. It is demonstrated that the temporal course of rivalry is sensitive to the presence of a subliminal moving stimulus within the currently suppressed field. The effects are seen to confirm a literal interpretation of Levelt's (1966) thesis which relates changes in the 'stimulus strength' of a rivalling field to subsequent changes in the temporal course of the phenomenon. This interpretation is consistent with the hypothesis that, despite phenomenal suppression, a full analysis is undertaken on the currently non-dominant stimulus. The data are related to models of selective attention, and to the notion that there are parallel visual systems.     

Measurement error in subliminal perception experiments: simulation analyses of two regression methods

A. G. Greenwald, M. R. Klinger, and E. S. Schuh (1995) investigated subliminal perception using a regression-based test for a dissociation between direct and indirect measures of perceptual ability. Direct and indirect measures were obtained for each observer, and a regression analysis was used to predict the amount of indirect perception at the point where the direct measure showed zero sensitivity. A significant positive intercept, obtained with both standard regression and a modified regression developed by K. C. Klauer, S. C. Draine, and A. G. Greenwald (1998), was used to argue for the key dissociation. When the assumptions of these methods are not met, however, simulations indicate that significant positive intercepts can often be obtained even if there is no dissociation. Moreover, the assumptions are theoretically implausible and inconsistent with some aspects of the results. Thus, the significant positive intercept is not strong evidence of the key dissociation.     

Perceptual vigilance and defense revisited: evidence against Blum's psychoanalytic theory of subliminal perception.

Blum's (1954) interpretation of psychoanalytic theory leads him to predict that Ss will defend against a threatening stimulus which is just below a recognition threshold and be vigilant toward the same stimulus when it is farther below the same threshold. 7 males and 9 females were presented the same four Blacky pictures, at the same speed and illumination, and using the same procedure as Blum. The results offered no support for the theory of perceptual vigilance or defense. The many weaknesses inherent in Blum's theory and supporting methodology were discussed.     

An examination of relationship between subliminal perception, visual information processing, levels of processing, and hemispheric asymmetries

A sublimiinal letter was exposed to the left or right hemisphere for either 15 or 30 msec. Subjective guesses were more accurate for visuo-spatial positional recognition made to presentations in the right hemisphere whereas verbal recognition was more accurate to presentations in the left hemisphere. The 30-msec. exposure increased the accuracy of the guesses. These findings were discussed in terms of differential triggering mechanisms for levels of hemispheric processing.     

Testing the theory of embodied cognition with subliminal words

In the current study, we tested the embodied cognition theory (ECT). The ECT postulates mandatory sensorimotor processing of words when accessing their meaning. We test that prediction by investigating whether invisible (i.e., subliminal) spatial words activate responses based on their long-term and short-term meaning. Masking of the words is used to prevent word visibility and intentional elaboration of the words’ semantic content. In this way, masking specifically isolates mandatory sensorimotor processing of words as predicted by the ECT. Do spatial subliminal words activate responses nonetheless? In Experiment 1, we demonstrate a spatial congruence effect of the invisible words if they precede visible target words. In Experiment 2, we show that masked words activate responses based on their long-term meaning. In Experiment 3, we demonstrate that masked words are also processed according to their short-term response meaning. We conclude that the ECT is supported by our findings and discuss implications of our results for embodied theories of semantic word processing and masked priming experiments.     

Conscious contributions to subliminal priming

Choice reaction times to visual stimuli (targets) may be influenced by preceding subliminal stimuli (primes). Some authors reported a straight priming effect i.e., responses were faster when primes and targets called for the same response than when they called for different responses. Others found the reversed pattern of results. Eimer and Schlaghecken [Eimer, M. & Schlaghecken, F. (2002). Links between conscious awareness and response inhibition: evidence from masked priming. Psychonomic Bulletin &Review, 9, 514-520.] showed recently that straight priming occurs whenever a prime is not efficiently masked thereby the information provided by the prime is accessible for consciousness. In the present study, a hypothesis is tested that straight priming is due to mediation of consciousness. To test this hypothesis, prime validity was manipulated. We showed that even when no mask was used so that participants could fully and consciously perceive the prime and participants were informed that primes were mostly invalid, for the short prime-target ISI interval (100 ms) straight priming occurred. The priming was inverse when the ISI was 800 ms. This indicates that participants were able to use the information provided by the prime to prepare the response opposite to that cued by the prime but only if the time between the prime and the target was long enough.