The Role of Task Conflict in the Non-Colour Word Stroop Task

The aim of the current thesis was to investigate the role of task conflict in the non-colour word Stroop task using the study-test procedure that was first introduced by MacLeod (1996) and then developed by Sharma (2018). Task conflict was suggested by two findings: (a) Studied words can slow down the colour-responding to unstudied words in a block with studied words compared to those in a block without studied words; (b) Within the studied block comprising of studied and unstudied words, the slowdown can also occur on a trial-by-trial basis when responses are made to two successively presented studied words - a reversed pattern of sequential modulation effect.

The thesis reports on three sets of manipulations and their effects on task conflict: the effect of using different (10 or 30) numbers of studied items (Chapter 4), the effect of varying the number of non-word rectangle stimuli (Chapter 5), and the effect of using emotionally salient words: anxiety-related and addiction-related words (Chapter 6).

Chapter 4 provided further evidence of colour naming interference from studied words and a reversed sequential modulation effect. There was a tendency that using 30 words produced less task conflict than using 10 words. Chapter 5 provided some evidence that task conflict increased with an increase in the proportion of rectangle stimuli. Chapter 6 again found evidence for task conflict indicated by interference from salient and/or studied words as well as a reversed sequential modulation effect.

The thesis explores how these findings can be explained by connectionist models of the Stroop task with particular emphasis on the proactive-control/task-conflict model by Kalanthroff et al. (2015). I also explore the role of individual differences and its role in top-down and bottom-up processes.