Switching attention and resolving interference: fMRI measures of executive functions
Introduction
Executive processes are responsible for controlling and coordinating the execution of goal-directed behavior [29]. In this paper, we are concerned with two situations in which such processes play a role. One is when there are multiple task goals, and attention must be shifted back and forth between the tasks based on the current task goal. The other is when there are two competing alternatives in some task, and the interference between the two must be resolved so that attention can be paid to one instead of the other. There is general agreement that both of these frequently discussed examples recruit executive processing (see [2], [31], [38], [42]). However, there is also a good deal of theoretical debate about the nature of executive processing. This debate has focused on the issue of whether there is there a single “central executive” process mediated by a single brain system, or there are multiple such processes, each different from the others in function and brain mechanism.
Several prominent theories of working memory and attention have promoted a singular view of executive function. According to this view, executive function can be conceptualized in terms of a unitary mechanism responsible for the allocation of attention to specific ongoing processes. An example of this view is Norman and Shallice’s [36] model of attentional control, which proposes that a unitary “supervisory attentional system” biases the activation of task schemas, favoring one over the others via inhibition or enhancement of activation values. Similarly, the framework for working memory introduced by Baddeley [1] proposed that it is a single “central executive” that manipulates the contents of a set of storage and rehearsal buffers in the service of some ongoing task. This singular view can be readily applied to the two situations that concern us here, as one might propose that both attention-shifting and response inhibition require only the allocation of attention; it is the allocation of attention to one task that inhibits the previously or currently irrelevant task.
By contrast, one can conceive of executive functions as a set of processes that are distinct from one another but that nonetheless work together in order to meet a particular common goal. In the task examples mentioned earlier, one might postulate one mechanism that allocates attention, another that coordinates the shifting of attention and information-flow between two tasks, and yet a different mechanism that resolves interference between the two tasks by inhibiting attention to the irrelevant one [40].
The behavioral literature on the issue of whether there are multiple or dissociable executive processes has led to mixed conclusions. Rogers and Monsell [39] demonstrated that the time taken to switch attention between two different tasks was disproportionately increased when interfering information was present, suggesting that the processes of attention-switching and interference-resolution interact with each other, and are not independent [44]. However, using a confirmatory factor analysis, Miyake et al. [35] concluded that the latent constructs of attention-shifting and interference resolution were only modestly related to one another, proposing that these processes may in fact be separable. Thus, the behavioral data suggest that attention-switching and interference resolution may share some common mechanisms, but may involve separable mechanisms as well.
If one assumes that different cognitive mechanisms are likely to be implemented in different brain systems, physiological data may provide information on the number and nature of executive functions. Whether tasks that involve each putative process activate different brain networks may provide insight into whether the processes are truly different, or whether they are simply competing conceptualizations of the same process.
Previous research using functional magnetic resonance imaging (fMRI) has examined processes of attention-switching and interference resolution in separate experiments using different tasks. In attention-switching paradigms, activation has been reported in dorsolateral prefrontal cortex (DLPFC) and posterior parietal areas in particular, although it is clear that these regions are part of a larger distributed neural network [15], [17], [25], [43]. With regard to tasks that require interference resolution such as the Stroop and Go/No-Go tasks, fMRI studies have shown activation in inferior and dorsolateral PFC, as well as anterior cingulate, again as part of a larger network of areas [10], [18], [22], [26], [28], [30], [32], [33], [41], [45].
In the current experiments, we used fMRI to determine if there are common or distinct areas of neural activation for the processes of attention-switching and interference resolution. We use a switching task similar to that used by Garavan and colleagues [16], [17] in which participants viewed a sequence of two stimulus types and were required to maintain internal counts for each type of stimulus. The task was constructed so that on successive trials, attention either remained on the previous count or switched to the other count. To study interference resolution, we used a task in which responses to the two stimulus types were either compatible with the stimuli, or required participants to inhibit the dominant compatible response and execute an incompatible one. In the first experiment we report, both tasks were combined in a single rapid event-related paradigm; in the second experiment, participants performed each task separately in a blocked design.
Section snippets
Participants
Fourteen undergraduate students ranging in age from 18 to 25 were recruited using advertisements in the university newspaper for Experiment 1. For Experiment 2, another 14 undergraduate students were recruited in the same manner. All participants were screened (using a self-report inventory) for neurological or psychiatric diagnoses as well as drug or alcohol abuse. They read and completed informed consent forms approved by the Institutional Review Board of the University of Michigan and were
Procedure
In this task, participants were presented with a sequence of centrally positioned arrows that pointed left or right. One of their tasks was to keep track of the numbers of left-facing and right-facing arrows in each block of 8–11 arrows. Participants were instructed to update the counts for both arrows silently after each arrow was presented, rehearsing first the count for the left arrow and then that for the right arrow, then making a motor response which initiated the display of the next
Experiment 2
Because the areas activated in Experiment 1 appeared close to FEFs and PEFs, a concern with the imaging data was that both experimental tasks might involve more overt or intended eye movements than their controls, and that mechanisms controlling these eye movements might produce activations that would interfere with the interpretation of the activations due to executive processes [11], [34]. To rule out this possibility, a saccade-control task was included in Experiment 2.
In addition, in
General discussion
The purpose of these experiments was to determine if common neural areas underlie the processes of attention-switching and interference resolution. Both of these processes are critically related to the ability to manipulate and control information in working memory, or what is often termed executive function. The results from the event-related analysis of Experiment 1, and the block design analysis of Experiment 2 showed similar areas of activation, providing a replication of findings in two
Acknowledgements
This research was supported by a grant from NIMH to the University of Michigan (MH-60655)
References (48)
- et al.
Dissociating working memory from task difficulty in human prefrontal cortex
Neuropsychologia
(1997) - et al.
Cortical activity during rotational and linear transformations
Neuropsychologia
(2000) - et al.
Direct comparison of prefrontal cortex regions engaged by working and long-term memory tasks
Neuroimage
(2001) - et al.
The role of frontopolar cortex in sub-goal processing during working memory
Neuroimage
(2002) - et al.
Differential activation of right superior parietal cortex and intraparietal sulcus by spatial and nonspatial attention
Neuroimage
(1998) - et al.
The effect of pacing of experimental stimuli on observed functional MRI activity
Neuroimage
(1997) - et al.
Prefrontal cortex activation in task switching: an event-related fMRI study
Cognitive Brain Research
(2000) - et al.
Thresholding of statistical maps in functional neuroimaging using the false discovery rate
Neuroimage
(2002) - et al.
Modulation of task-related neural activity in task-switching: an fMRI study
Cognitive Brain Research
(2000) Executive control of thought and action
Acta Psychologia
(1985)