In the dichotic listening task, people are asked to listen through a set of headphones to two different messages that are being played to each ear and are asked to focus on the information that is being played to one ear. An experiment done by Cherry in 1953 showed that people can never tell very much about the information that was going into their unattended ear aside from base characteristics such as the sex of the speaker and whether or not it was a human voice. Another experiment by Kotecki, Brehm, and Healey (2002) tested this further by playing the same message into the unattended ear as into the attended ear at a five second delay. It was shown that the vast majority recognized no similarity between the two messages when questioned later. This effect was also noted by Treisman (1964). This effect has been compared to a cocktail party, in which people are able to engage in conversations with others in a crowded room with little difficulty. However, when a person's name is said, they almost inevitably distracted and turn towards the speaker. (Anderson, 76)
An experiment by Glucksberg and Cowan (1970) had people shadowing a message in one ear, with a randomly said digit going into the message of the other ear. If asked within five seconds of hearing the digit, people were able to recall that a digit had been said about 25% of the time. This shows that there is some processing of the information in the unattended ear is occurring. An experiment by Kotecki, Healey, and Brehm was done in which participants shadowed both English and French, with strong language and sexual content being played into the unattended right ear. The group found that people were more likely to be distracted during the French segment except if the participant had a fairly strong background in a foreign language, especially French. (2002) This experiment has been designed to eliminate the possibility of language familiarity to see if the same effect will still occur. It was hypothesized that the loss of familiarity and context would result in the participant being more easily distracted by the information in the unattended ear.
The seventeen participants for this experiment were all college age people, either recent graduates or current students of a small private college. The experiment was run using a Macintosh computer that used the SoundEdit program. Participants wore adjustable headphones for the dichotic listening task. The data was recorded on a piece of paper upon which the entirety of the unattended message was typed, any of the words or phrases that were said were underlined upon being said by the participant. The stimuli consisted of two messages. The first, which was played into the left speaker of the headphones, consisted of two parts. The first was a reading from Stephen King's "The Shining", which was read by a male voice for approximately thirty-five seconds. The second part was made entirely of gibberish, which was read by the same male voice for the same amount of time. The message that was played into the right unattended ear was a different male voice that began five seconds after the left ear message had begun and ended at the same time as the attended message. This consisted of random sentences selected from various "Playboy" magazines that had been selected for explicit, attention-grabbing content. The sentences were then altered and arranged by J. Brehm from this group into a coherent paragraph, which was consistently dirty through the entire thing. The participants were first given a practice run to familiarize them with shadowing. They shadowed an excerpt from "The Tin Drum" by Gunter Grass. Afterwards they were told to shadow their left ear, and were warned of the switch between English and gibberish. They were also asked to quickly repeat any words or phrases that caught their attention in their right ear, and to then go back to the shadowing task. The distracting words or phrases were underlined on a typewritten piece of paper as the participant indicated occurring distraction.
It was found that eight of the participants were more distracted during the English portion and eight were more distracted during the gibberish, leaving one participant who was equally distracted during both segments. When comparing instances of distraction, it was found that more occurred during the gibberish portion, but not significantly so (see Figure 1).
Figure 1. Comparing the number of instances in which participants were distracted. As is shown on the graph, more instances in which the unattended message broke through were recorded during the gibberish portion, but not significantly.
However, when trying to incorporate the fact that sometimes the participants were distracted for an entire phrase, the number of words was counted for each section. This time more distraction appears to have occurred during the English portion of the attended recording, but once again there was not a large enough difference to be significant (see Figure 2).
Figure 2. Comparing the number of words that were distracting during the two segments. When the number of words is counted instead of just instances of interruption, the attended message that was recorded in English shows itself to be slightly stronger, but not significantly.
Even though there were no significant differences between the susceptibility to distraction for the two portions of the attended message, some conclusions can be drawn when compared to previous experiments. For instance, as was mentioned earlier, there was Kotecki, Brehm, and Healey's 2002 experiment where participants had to shadow both English and French. If for that experiment it was found that people had more difficulty shadowing French, when compared with this experiment the difference is that the gibberish was not an actual language. Even though in both cases the participants were told to do their best phonetically, the fact that one was a real language may have led the participants to struggle more with getting the actual words correct, while with the gibberish they were just repeating the sounds that they heard, in which case there was no pressure to be correct. This may have led the participants to focus more on the attended ear, leading to less distraction.
In the case of the experiment where the participants were shadowing French, knowing it was a true language, the English in the unattended ear might have more easily distracted them. Also, the pronunciations were more difficult when attempting to replicate French, while the gibberish was essentially nonsense English. This leads to the notion that while participants are not really getting much context out of the unattended ear, they may not even be really aware of what they are saying. It would seem that this might be necessary while shadowing, so that less confusion can occur.
Anderson, J.R. (1995) Cognitive Theory and its Implications Fourth Edition. W.H Freeman and Company
Cherry, E.C. (1953) Some Experiments on the Recognition of Speech, with One and with Two Ears. The Journal of the Acoustical Society of America. 25 (5), 975-979
Kotecki, E., Brehm, J., and Healey, R. (2002) Distraction of the Echoic Memory. Unpublished paper.
Kotecki, E., Brehm, J., and Healey, R. (2002) Echoic Recognition of Similar Readings. Unpublished paper.
Montague, W.E. (1965) Effect of Irrelevant Information on a Complex Auditory-Discrimination Task. Experimental Psychology. 69 (3), 230-236
Treisman, A. (1964) Monitoring and Storage of Irrelevant Messages in Selective Attention. Journal of Verbal Learning and Behavior. 3, 449-459
More Posters
Web Posters from other courses and projects
Alma College Psychology Department
Source of Distraction