Advertisement

In sight but out of mind: Do competing views test the limits of perception without awareness?

Description
In sight but out of mind: Do competing views test the limits of perception without awareness?
Categories
Published
of 9
2
Categories
Published
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Similar Documents
Share
Transcript
  In sight but out of mind: Do competing views test the limitsof perception without awareness? q Matthew Brown * and Derek Besner * Department of Psychology, University of Waterloo, Waterloo, Ont., Canada N2L 3G1 Received 22 September 2003 Available online 22 January 2004 Abstract Over a century  s worth of research suggests that ‘‘perception’’ without awareness is a genuine phenom-enon. However, relatively little research has explored the question of whether  all   visually presented infor-mationactivatesrepresentationsinlongtermmemorywithoutawareness.Twoexperimentsexploredtheuseof a figure–ground display consisting of competing views in which one view dominates the other such thatsubjects are (initially) unaware of the non-dominant view. Neither experiment provided evidence that thenon-dominant view (an embedded word) activated its representation in long term memory when the subjectfailed to report being aware of the embedded word in that priming was not seen on a subsequent stemcompletiontask.Incontrast,primingwasseenwhensubjectsreportedbeingawareoftheembeddedword.Itis suggested that two competing figure–ground relations are not concurrently computed unconsciously.   2003 Elsevier Inc. All rights reserved. Keywords:  Perception without awareness 1. Introduction The question of whether ‘‘perception’’ occurs in the absence of awareness has intrigued laymenand psychologists alike for over a century, and has generated many attempts to prove that itis possible (e.g., Sidis, 1898). A more recent demonstration of perception without awarenessis Marcel  s (1980) semantic priming experiments where subjects were presented with three q This work was supported by an Ontario Graduate Scholarship to M.B. and Grant AO998 from the NaturalSciences and Engineering Research Council of Canada to D.B. * Corresponding authors. Fax: +1-519-746-8631. E-mail addresses:  ms2brown@watarts.uwaterloo.ca (M. Brown), dbesner@watarts.uwaterloo.ca (D. Besner).1053-8100/$ - see front matter    2003 Elsevier Inc. All rights reserved.doi:10.1016/j.concog.2003.11.004 www.elsevier.com/locate/concog ConsciousnessandCognition Consciousness and Cognition 13 (2004) 421–429  consecutive letter strings and asked to make a lexical decision to the first and last string. On thecritical trials, the second string was a polysemous word (e.g.,  palm , referring to either a type of treeor a part of the hand) and was either pattern masked or not masked. The word triad was eithercongruent (e.g.,  hand   –   palm  –  wrist ) or incongruent (e.g.,  tree  –   palm  –  wrist ). Marcel found that whenthe polysemous word was not masked, semantic priming was observed for congruent triads, butnot for incongruent ones. In contrast, when the polysemous word was masked so that subjectswere unaware of it, semantic priming was now observed for both congruent and incongruenttriads. This finding suggests both meanings of a polysemous word are activated without awarenessprior to the appropriate interpretation (as defined by the context) being selected such that thesubject will become aware of it.Marcel  s (1980) findings, in conjunction with many other subsequent studies reporting similarresults over the last two decades, has led to the conclusion that perception without awareness is agenuine phenomenon (e.g., see Merikle, Smilek, & Eastwood  s, 2001 review). Traditionally,subjects are rendered unaware of visual information either by (a) masking the stimulus, or (b)requiring subjects to orient their attention away from the stimulus (e.g., Mack & Rocks, 1998inattentional blindness paradigm). That said, few studies have attempted to establish whetherboundary conditions exist that limit the kinds of information that can be concurrently activatedwithout awareness. One possibility is that although multiple meanings of a word can both beunconsciously active, the visual system might only be able to  unconsciously  activate a single fig-ure–ground relationship at a time.The Necker cube (Necker, 1832) provides an example of visual information that contains twodifferent low-level visual perspectives. In theory, both of these perspectives could concurrently beunconsciously active. In the standard Necker cube display (Fig. 1A), a three-dimensional cube canbe viewed from two competing visual perspectives. In the first perspective (Fig. 1B) the outer topis shaded in gray, whereas in the second perspective (Fig. 1C) the outer bottom is shaded in gray.Although both perspectives are visible in Fig. 1A, they are never perceived simultaneously, butrather alternate over time.If a Necker cube display is presented briefly and then removed such that subjects are only awareof one of the two competing perspectives, what is the fate of the yet to be consciously seen otherperspective? If multiple lower-order visual perspectives (e.g., two competing views of the Neckercube) can be concurrently and unconsciously active, then both views should produce priming on ameasure sensitive to this kind of activation. A finding of this nature would be analogous toMarcel  s (1980) result that both congruent and incongruent triads produced semantic primingwhen subjects were unaware of the second (polysemous) word of the triad. If both visual Fig. 1. (A) The standard Necker cube with two competing perspectives. (B) Competing perspective one: the outer topshaded in gray. (C) Competing perspective two: the outer bottom shaded in gray.422  M. Brown, D. Besner / Consciousness and Cognition 13 (2004) 421–429  perspectives cannot be unconsciously active at the same time, then the yet to be consciously seenview should not produce priming.The Necker cube display does not seem particularly well suited to experimental investigationsof perception without awareness (at least by us) given the degree of visual overlap between the twocompeting views. This overlap makes it difficult to devise a measure of perception withoutawareness for the   unseen   perspective without that measure being contaminated by the perceptionof the   seen   perspective. The present experiments solve this inherent limitation of the Necker cubeby simply using a different kind of stimulus.Fig. 2 illustrates a display in which two lower-order visual perspectives compete for consciousperception. This display is constructed such that one perspective, the black geometrical blocks, ismore visually salient than the other, the embedded word CLOTH. As such, a brief presentation of this figure typically results in the conscious experience of only seeing black geometrical blocks.The question addressed here is whether the embedded word, despite not being consciously per-ceived, nevertheless activates its unconscious lexical-semantic representation in memory. Priorwork with the stem completion task has successfully demonstrated perception without awareness(e.g., Mack & Rock, 1998; Merikle, Joordens, & Stolz, 1995). This prior work used distribution of attention and visual masking to render observers unaware of a stimulus. If the embedded word inthe present paradigm activates its representation unconsciously, then it should produce primingon a subsequent stem completion task where subjects are asked to complete a stem ‘‘with anyletters that come to mind that make a word.’’To foreshadow the results, the present experiments provide no evidence that the consciously  unseen   embedded word is processed without awareness. We take these results up further in thediscussion section. 2. Experiment 1  2.1. Method  2.1.1. Participants Fifty-nine University of Waterloo undergraduate students participated. All subjects spokeEnglish as a first language and were assumed to have normal or corrected-to-normal vision.Subjects were paid for their participation.  2.1.2. Design The experiment had two conditions: an experimental and a control condition with 25 and 34subjects in each, respectively. Subjects in the experimental condition saw either the  non-embedded Fig. 2. An example of a figure–ground display with the embedded word   CLOTH.  M. Brown, D. Besner / Consciousness and Cognition 13 (2004) 421–429  423  word ‘‘CLOTH’’ or ‘‘STUDY,’’ followed by a presentation of the other word in  embedded  form.The purpose of the non-embedded word was to confirm that, under the conditions of the presentexperiment, if the subjects were aware of a word (i.e., reported seeing it), they would use it tocomplete a subsequent stem. These subjects then completed two separate word stems in succes-sion, where the first word stem always consisted of the first three letters of the embedded word andthe second word stem consisted of the first three letters of the non-embedded word. Subjects in thecontrol condition were only asked to complete the two word stems (i.e., the embedded and non-embedded word displays were not presented). The purpose of the control condition was to pro-vide an estimate of the baseline probability of subjects completing the stems with the embeddedwords that appeared in the figure–ground display (i.e., either CLOTH or STUDY).  2.1.3. Stimulus materials The two words 1 (i.e., CLOTH and STUDY) were selected from Jacoby (1998), and werematched in terms of the number of possible stem completions (eight) and frequency (high fre-quency). The letters of the embedded word appeared in white whereas the spaces between themand at the beginning and end of the embedded word were presented in black (see Fig. 2). The non-embedded word was created by reversing the colors used in the figure–ground display. That is, theletters were presented in black and the spaces between the letters and at the beginning and end of the word appeared in white.  2.1.4. Procedure Subjects were tested individually and were seated approximately 57cm from the computermonitor. Subjects in the experimental condition were instructed to pay close attention to themonitor since two consecutive displays (i.e., the non-embedded word and the embedded word)would appear only briefly. All subjects were given stem completion instructions. Subjects thenpressed the spacebar to initiate the presentation of the two displays. Subjects in the experimentalcondition were shown the non-embedded word for 500ms and then the figure–ground display for500ms, separated by a 1000ms inter-stimulus interval. The stem corresponding to the embeddedword was presented immediately after the offset of the figure–ground display. Subjects completedthe stem (verbally), which was recorded by the experimenter. Subjects were then asked to reportwhat they had seen prior to the presentation of the stem, which was also recorded by the ex-perimenter. Subjects then completed the stem corresponding to the non-embedded word. Theprocedure for subjects in the control condition was identical to that described above, except thatthey only participated in the stem completion phase of the experiment.Stimuli were displayed on an ADI Micro Scan color monitor controlled by Micro ExperimentalLab (MEL) software (Schneider, 1990) implemented on a Vault PC compatible computer. Thefigure–ground display subtended a visual angle of approximately 29.2   horizontally and 6.4  vertically. The non-embedded word display subtended a visual angle of approximately 25.2  horizontally and 6.4   vertically. 1 These words were selected with the constraint that they did not contain any diagonal features (e.g., the feature thatdifferentiates a   P   from an   R  ). Diagonal features appear to be visually salient, resulting in the embedded wordbecoming easily visible.424  M. Brown, D. Besner / Consciousness and Cognition 13 (2004) 421–429   2.2. Results and discussion Thedatafromthreeofthe25subjectsintheexperimentalconditionwerediscardedbecausethesesubjects reported seeing the embedded word. Given that the purpose of this experiment is tomeasure perception without awareness, these data are of no interest. Three subjects could notprovide an acceptable completion for one of the two stems (one for the stem corresponding to theembedded word and two for the stem corresponding to the non-embedded word) but were never-theless included in the analysis because they provided an acceptable completion for the other stem.  2.3. Embedded word  The data from the remaining 21 subjects in the experimental condition and the data from 34subjects in the control condition were then submitted to the   observed   and   expected   columns,respectively, of a one-way  v 2 test (see Table 1). The control condition data were obtained bycalculating the proportion of subjects that did and did not use the embedded word to complete thestem. However, given that proportions are not legitimate entries that can be used to compute  v 2 (see Howell, 1999, p. 384), they were converted to frequencies that summed to equal the frequencytotal in the experimental condition. The results from this analysis yielded no significant differencebetween the observed data and the expected (i.e., control) data,  v 2 ð 1 Þ ¼  0 : 31,  p  >: 50. That is, thepresentation of the figure–ground display did not result in the embedded word being used abovechance to complete a word stem for those subjects who failed to report seeing an embedded word. Table 1Number of subjects as a function of Experiment (1 vs. 2), report of embedded word (Yes vs. No), display type (Em-bedded vs. Non-embedded), inclusion on stem completion task (Yes vs. No), and condition (Experimental vs. Control) Experiment Report seeingembeddedword?Display type Used tocompletestem?ConditionExperimental(observed)Control a (expected)1 No Embedded Yes  4 5.25 No  17 15.75 Non-embedded Yes  12 5 No 8 152 No Embedded Yes  7 5.75 No  16 17.25 Non-embedded Yes  18 5.5 No 4 16.5Yes Embedded Yes  13 4 No  3 12 Non-embedded Yes  12 4 No 4 12 a These values represent data from the control condition that were transformed with the constraints that they (a) sumto total the number of subjects in the experimental condition and (b) maintain the ratio of    yes   to   no   responsesobserved in the control condition. M. Brown, D. Besner / Consciousness and Cognition 13 (2004) 421–429  425
Search
Related Search
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks
SAVE OUR EARTH

We need your sign to support Project to invent "SMART AND CONTROLLABLE REFLECTIVE BALLOONS" to cover the Sun and Save Our Earth.

More details...

Sign Now!

We are very appreciated for your Prompt Action!

x