Competing responses and the processing of irrelevant information

Abstract

A number of studies have shown that performance of identification and discrimination tasks is detrimentally affected by irrelevant information, yet other studies have failed to find such decrements. It is suggested that these contradictory findings depend on whether S must make difficult discriminations among the relevant stimuli, the irrelevant stimuli, or between the relevant and irrelevant stimuli. The role of irrelevant information in these tasks is to enhance or amplify the competing responses engendered by the difficult discriminations. Irrelevant information enhances competing responses by increasing the information processing requirements of a task. The results of various studies of irrelevant information seem to be in good agreement with the assertions.

Processing nominal information during visual search.

“…Indeed, Henderson makes no attempt to specify the 'level' of abstraction of the name code, and it will, therefore, be argued that naming did not take place until some level of structural representation had been achieved during the second stage of analysis. Under these circumstances, (a) the naming of the target's other case when it was present as an irrelevant item induced a competing response (see Hodge, 1973), thereby delaying its rejection as a non-target, and (b) the absence of response competition when the two different case targets shared a name facilitated the execution of the response. This is likely if we accept Neisser's (1967) description of visual search, and is supported by some aspects of Henderson's study.…”

“…Thus, Neisser considered initial processing to be prone to error, and it must follow that in any search task where subjects make few errors, some irrelevent items are passing the first stage and being rejected during or after detailed analysis. Also, since subjects searched for all targets during any one experimental session, it might be expected that the probability of selecting one of these 'sometimes-relevant' irrelevant distractor items would be greater than the probability of selecting a 'never-relevant' irrelevant item (see Hodge, 1959Hodge, , 1973. Consequently, (a) when the target's other case was used as a distractor, and on those occasions when it was incorrectly selected and named, delays in response would be quite likely, and (b) when search was for the target and its other case, the correct selection and naming of either letter would be more likely to facilitate a response than the correct selection and naming of either of two different case targets not sharing a name.…”

The effects of amount of information in the Stroop color word test

“…While there is no evidence in the present study to indicate that response competition does exist, failure to find such evidence does not rule out the possibility that response competition of some sort can and does take place. Numerous studies on the Stroop task and related phenomena (e.g., Dalrymple-Alford & Azkoul, 1972;Egeth, Blecker, & Kamlet, 1969;Hodge, 1973) indicate that interference is dependent on the type of response required. Combining the results of these studies with the present findings, supporting Stroop interference at the perceptual encoding stage, a combined perceptual and response competition approach is suggested.…”