Sign In Sign-Up

Welcome!

Close

Would you like to make this site your homepage? It's fast and easy...

Yes, Please make this my home page!

No Thanks

  Don't show this to me again.

Close

Introduction:

The aim of the present study was to examine the possibility of priming during the period of the attentional blink. In almost all attentional blink studies, it has been found that the recall of T2 is hampered by the identification of T1. With this in mind the present study examined the possibility that T2 was not completely rejected from attentional processing, but is instead processed to the level of meaning. The present study examined the role of unconscious processes operating during the period of the attentional blink, employing a different measure of priming from that used by Shapiro et al (1997) and Maki et al (1997). It was decided that a stem completion task would be a good measure of priming. This decision was based on the findings of Rajaram and Roediger III (1992) in which they found stem completion was a better measure of implicit memory than words presented in the auditory modality; words presented in same typeface and words presented with different typefaces. Rajaram and Roediger III found word stem completion to be an effective measure of priming thus unconscious processing. In the present study, T2 was always presented within the RSVP stream after the T1 and T1 was either present or absent, thus an attentional blink should only be produced for those T1 present trials. Therefore priming may reflect a difference in the two conditions. It is hypothesised that participants will significantly report fewer incidents of T2 in the presence of T1 than in the absence and that this will be reflected in the level of correct priming of T2.

Method:

Participants: Top of page

Six postgraduate (one male, five female) ranging in age from 22 to 38 (M = 23.5 years) volunteered to participants in the study. All subjects were naïve to the experimental procedure, and signed an informed consent form (see appendix one) prior to taking part in the study. Subjects were made aware of the fast flashing stimuli and the slight possibility that it may trigger an epileptic seizure. All subjects reported having normal to corrected to normal visual acuity, and no deficits in colour vision.

Apparatus: Top of page

The RSVP program was generated by an IBM personal computer using custom made software, and was displayed on an a 28.6 centimetre (cm) (13 inches, viewable area) colour monitor. Subjects were asked to view the computer screen from approximately 35-cm (20 inches). Responses were recorded by the computer and were later subjected to statistical analysis. An initial response sheet contained 60 word stems that corresponded to the second target item, which subjects completed before taking part in the second experimental procedure. Each word fragment made up a five-letter word, and each word was drawn from a list of high frequency English words as described Friendly, Franklin & Hoffman (1980).

Stimuli: Top of page

These consisted of a RSVP stream of 13 black 5-lettered words, 1 white T1 word and 1 blue T2 word; making each stimulus stream a total of 15 words long. All words were approximately 16mm wide and 4mm high, uppercase and presented in Courier font. The words presented in the RSVP stream were chosen from the Toronto words pool (Friendly, Franklin & Hoffman, 1980), and had a logarithm frequency of between 1.10 and 6.57. Each word was presented for 65 ms with an inter-stimulus interval (ISI) of 15 ms, and displayed at the same location in the centre of a uniform blue-grey screen (see appendix two).

Design: Top of page

A repeated measures design was utilised, such that each participant took part in all conditions of the independent variable, namely: 'T1 present' trials (experimental condition) versus 'T1 absent' trials (control condition). T2 detection was the dependent variable

Procedure: Top of page

This initial study involved asking volunteers (N=6) to first complete a stem completion task comprising of 60 word fragments, (see appendix three) after a time interval of no less than twenty-four hours, subjects participated in the second experimental condition in which they took part in 60 RSVP trials. This consisted of 30 trials in which the T1 was present and 30 trials in which the T1 was absent with T2 present in all 60 trials. The T1 always-appeared in serial position 5 and the T2 always-appeared in serial position 8. The word 'TABLE' was used as the T1 word in all randomly generated 'T1 present' trials. In addition, a single T2 word was chosen from a list of 60 other distractor words for each trial. The computer randomly generated the position of 13 other distractor words that made up the full RSVP stream for all 60 trials. Each participant initiated a trial by pressing the space bar on the keyboard. Each trial began with a 180 ms presentation of a small, black fixation dot. The participant's first task was to report the presence or the absence of the white word (T1) at the end of every trial (e.g. "Did you see the word TABLE?"). Participants were told what the target word was, and not to respond to its colour alone. The participant's second task was to complete word stems for all T2's at the end of every trial. For example, "Please complete the following word stem, MU_ _ ", etc. The stem would correspond to T2 in each trial. For example if the second target was 'MUSIC' and the stem completion task would look like MU_ _. The computer generated the relevant questions for each task for all trials. All responses (i.e. 'yes' or 'no') were keyed into the computer and later printed out. Participants were told to press the 'enter key' if they were unsure of any correct response. No time limit was set for each experimental session, but none took longer than 15 minutes.

Results: Top of page

Raw data was collected for both T1 present and absent (see appendix four), in which in all participants completed word stems for those trials in which the target was correctly detected as being present or absent. The data was analysed in a number of ways; firstly the stem completion results were divided into stems that matched the expected completions and stems that mismatched the expected completions. The stems that matched the expected T2 words were disregarded from any further statistical analysis. The matched stems indicated that subjects were able to guess correctly the expected words without the need to be primed. The mismatched stems were then compared to the corresponding T2 items, if the subject gave an incorrect stem on the stem-completion task but then had given the expected stem on the second task, this counted as a positive result, seemingly indicating a priming effect. However, if the subject had given an unexpected stem on both tasks the results was disregarded. On several occasions in the initial stem-completion task a number of subjects failed to complete stem words. If on the second RSVP trial the subject then completed the stem as expected, a positive result was recorded; this would also indicate a priming effect. One of the last tasks to be performed was to divide the end results into two categories, T1 present (experimental condition) and T1 absent (control condition). Table One shows the means, standard deviation and the standard error calculated for the raw data, the table of results shows that the mean number of word stems completed as expected for T1 absent and T1 present trials. The results were subjected to a correlated t-test, (t = 0.806) indicating no significant difference between the means of the two conditions, T1 absent and present. This finding appears to suggest the absence or attenuation of the attentional blink, in the presence of T1, that according to previous findings should have interfered with the word stem completion. The results therefore do not support the hypothesis, which states that participants will significantly complete more stems in the absence of T1 then in its presence. These results indicate that the attentional blink does not prevent implicit processing from occurring; thus, indicating that during the hypothesised attentional blink the attentional system does not momentarily close down, as has previously been hypothesised.

Conclusion / Discussion: Top of page

The results of this experiment calculated no significant difference between the means of stem-completion words when T1 was absent or T1 present. Therefore, participants did not differ significantly in the number words stems completed in the presence or absence of T1. Therefore the experimental hypothesises was rejected, in favour of the null hypotheses, namely that there would be no significant difference in the mean results in the absence or presence of T1. This result appears to indicate that the attentional blink does not prevent participants from successfully completing the expected word stem. This therefore suggests that participants were primed to the expected stem, in the presence of T1, which according to previous findings should have interfered with the priming of T2. There are however, a number of problems with the experimental design that may account for the results found, and may lead to a different conclusion being drawn from the one stated above. From the initial analysis of the stem-completion task, it was clear that many stems were far too easy, and that many had a limited number of possible completions. For these reasons the vast majority of stems were correctly completed without first being primed. In most cases the subjects only mismatched the expected stem 25% of the time and participant one only mismatched the expected stem on four occasions. This may highlight a fundamental difficulty with the experimental design. The stem completions had a high frequency rate in the English language. A possible follow up experiment may include using words with a lower word frequency rate. However, this may limit the number of possible word completions. The experimental design also allowed the subjects to leave unfinished stems, which they did not know. In a follow on studies subjects should not be given the choice to leave any unfinished stems, thus incorporating a forced choice paradigm to the study. A way to correct the possible design errors would be to extend the trial period. In many other attentional blink studies subjects are asked to attend to more than 300 RSVP trials. In a possible follow on study a trial period comparable to main stream attentional blink studies, may be needed. This would have the effect of being able to introduce a number of different elements. In an extended trial period word frequency could be altered, to low and high frequency words. Also, incorporating, a design element where the stem completion in the RSVP element of the study, gave a mismatch between T2 and the stem. This may highlight whether participants are being primed to the stem are or simply guessing the completion. Even with the altered design elements it is suspected that a word stem-completion paradigm, is not robust enough for an attentional blink study.


References: Top of page

Maki, S.W., Frigen, K., & Paulson, K (1997) Associate Priming by Targets and Distracters During Rapid Serial Visual Presentation: Does Word Meaning Survive the Attentional Blink. Journal of Experimental Psychology: Human Perception and Performance Vol 23 No 4. 1014-1034.

Rajarm, S., & Roediger, H. L. (1992). Direct comparison of four implicit memory tests. Journal of Experiment Psychology: Learning, Memory and Cognition, 19 (4), 765-776.

Shapiro, K., Driver, J., Ward. R., & Sorensen. R. E. (1997). Priming from the attentional blink: A failure to extract visual tokens but not visual types. Psychological Science, Vol 8, March

Appendix One

The following study is an experiment into learning. The stimuli are produced on a computer screen in a rapid serial visual presentation. If you feel that you may suffer adverse affects to the fast flickering stimuli please excuse yourself from the study now. Also, if you are dyslexic, suffer migraines or have been knocked unconscious at any time in your life, you should withdraw from the study. Furthermore, as with all psychological experiments you have the right to withdraw from the study at any stage. I have read and understood the statement above and wish to participant as a volunteer in the study. Participant'

Signature…………………………….Date……/……../2000

Name…………………………………

Experimenter' Signature…………………………………………………………..

Please leave your email address here…………………………………..