The Experiment Aims to Test the Stroop Effect

Klein (1964) found that there was an interference gradient, with the interference decreasing as the colour-relatedness of the word decreased when testing colour-naming using semantic variations. This experiment aims to test the Stroop effect in colour-naming, using the same six semantic variations as Klein (1964) to determine if there is an interference gradient. Participants were asked to name the colour of the ink of 80 words, with different levels of colour-relatedness, in each of the six conditions. Each participant only took part in one of the six conditions. The experiment found that there was an interference gradient with the classic Stroop condition having the highest time increment, with the time increments then decreasing across the conditions as the colour-relatedness of the words decreased, supporting Klein’s (1964) findings. This is important as it suggests that there may be semantic interference in colour-naming Stroop tasks. Introduction Stroop (1935) found that, when participants were only reading the colour words, they could largely ignore the ink colour. Stroop (1935) also found interference in colour naming when the colour words were incongruent.

This is known as the Stroop effect, the interference that the colour-word has on the speed of naming the colour. The Stroop effect has been studied in various ways. MacLeod’s (1991) review of studies on the Stroop effect shows that around 300 studies, of the over 700 conducted, have looked at interference as a measure of attention. In such studies there are numerous variations that have been studied such as Stroop colour-word interference tests (Stroop 1935; McCown & Arnoult 1981; Regan, 1978), picture-word interference tasks (Rosinski, Golinkoff & Kukish, 1975; Rosinski, 1977; Rayner & Springer, 1986), acoustic variation (Bakan & Alperson, 1967; Underwood, Briggs, & Underwood, 1984), semantic variation (Klein, 1964; Murray, Mastronardi & Duncan, 1972; Darlymple-Alford, 1972) and many more variations. This experiment will focus on semantic variations in the Stroop task. Klein (1964) investigated the Stroop effect using a semantic variation task. Klein (1964) found that the amount of reading delay, when naming colours, depended on how related the ‘vocal components’ (Klein, 1964, pg 587) and the colour-relatedness of the words were to the correct colour-response. Klein’s (1964) experiment consisted of six conditions: nonsense syllables; rare words; common words, not closely associated with colour; colour related meanings; colour names (distant); colour-names (close).

In each condition, text was coloured red, green, yellow and blue and participants were asked to name the colour of the word. Klein’s (1964) findings were replicated by Darlymple-Alford (1972) who found that colour words interfered more than colour-related in comparison to unrelated control words. Klein’s (1964) findings are also supported by the findings of Murray, Mastronardi, & Duncan (1972) who found that subjects were able to name the colour of animal words faster than colour-words. Klein’s conclusions were also replicated by Fox, Shore & Steinman (1971) who found that the reading speed for ‘common’ neutral words was significantly slower than those for ‘rare’ neutral words. Klein (1964) gives several explanations for his semantic interference gradient, related to verbal context. One is the attensive power of the word, its ability to produce arousal. The greater this power, the more interference the word will produce (Klein, 1964).

MacLeod (1991) gives the explanation of automaticity, that naming the colour uses more attentional resources than reading the irrelevant word, and so colour naming takes longer. Automaticity also suggests that reading is more automatic than colour naming (MacLeod, 1991). The more automatic processing is believed to interfere with the less automatic process of colour naming (MacLeod, 1991). Hock & Egeth (1970) suggest an explanation of perceptual encoding. This explanation suggests that perceptual encoding of the colour of the ink is slowed down by the opposing information received from a colour word. From the results of their experiment they also suggested that colour-related words are recognised faster than less colour-related words and so are more likely to interfere with encoding the colour of the word (Hock & Egeth, 1970; McLeod, 1991). Cohen, Dunbar & McClelland (1990) devised a Parallel Distributed Processing Model as an explanation for the Stroop effect, which integrates elements of automaticity and processing speed into one framework. Processing occurs when the system is activated by a stimulus. This processing continues along pathways of differing strengths, flowing in one direction – from input to output. Pathways are comprised of interconnected modules containing units, with processing pathways, including some or all of the units in various modules, being selected when activation occurs.

The set of connections specifies the strength, determining the speed and accuracy of processing. Interference occurs if two pathways are activated simultaneously that produce conflicting activation (Cohen, Dunbar & McClelland, 1990; MacLeod, 1991). This experiment will be a partial replication of the experiment conducted by Klein (1964). This experiment will use the same six experimental conditions as Klein: colour words (classic Strrop); other colour words; high-frequency colour-related words; high frequency ‘neutral’ words; low frequency ‘neutral’ words; letter strings (this uses the same stimuli as nonsense syllables). This experiment will also measure the time taken to name the ink colour in the six experimental conditions but will use orange ink instead of yellow ink. This experiment will also use a mixed design instead of using a between-participants design, used by Klein (1964). There are two main predictions for this experiment. The first is that, in the two control conditions, reading will be faster than colour-naming and so condition A will be faster when compared to condition B. The second is that there will be a graded interference effect across the six experimental conditions. This means that the classic Stroop condition will have the highest mean time increment of all six conditions with the time increment then decreasing across the rest of the conditions as the colour-relatedness of the words decreases, being the lowest for the letter strings condition.