Original Article
Objective
:
To
explore the neural correlates of the retrieval of previously acquired
information during top-down control.
Methods
: T2*-weighted echoplanar imaging sequences (24 axial slices, flip angle/TR/TE=90º/2500ms/TE=50ms) obtained with a GE 1.5 T scanner were used to collect 120 blood oxygen level-dependent images from ten healthy volunteers performing the modified Stroop task. The imaging data were investigated to determine, firstly, the network used to solve the Stroop interference test with previously acquired information and, secondly, the neural correlates of the retrieval of the previously acquired information. Activation maps were acquired by means of the random effect model using SPM2.
Results
: More extensive regions, including the right middle frontal gyrus, right precuneus, left cerebellum, right supplement motor area, left insula, and left middle temporal and occipital gyri, were needed to solve the Stroop task with previously acquired information than to solve the Stroop-only task. The left precuneus and right superior temporal gyrus (STG) were among those regions which remained activated in the Stroop condition with previ-ously acquired information after using the Stroop mask.
Conclusions
: This study demonstrated that the left precuneus and right STG are neuronal correlates of the retrieval of previously acquired information during top-down control. Further studies using connectivity analysis are needed to clarify the top-down control mechanism using previously acquired information.
Correspondence: Bum Seok Jeong, M.D., Ph.D., Department of Neuropsychiatry, Eulji University School of Medicine, 1309, Dunsan-dong, Seo-gu, Daejeon 302-799, Korea
Tel: +82-42-611- 3443, Fax: +82-42-611-3445, E-mail: drshrink@eulji.ac.kr
The color word Stroop task is a widely used measure of the top-down control of behavior.1,2,3 The Stroop effect is said to occur when people are asked to say the color of a word, which is the name of a different color, and the overlearned word inhibits them from doing this. Recent functional neuroimaging studies using Stroop task have shown that the function of the anterior cingulate (ACC) is related to conflict monitoring,4,5 error detection,6 and response selection,7 while the function of the dorsolateral prefrontal cortex is related to strategic processes.8,9 In daily living, however, a previously acquired experience is needed to achieve the top-down control of behavior, or alternatively information as well as an overlearned one being used during the Stroop task. For example, if a driver has to choose which way to go at a fork in the road, his or her judgment will be based on past experience or a signpost previously encountered. To perform this operation successfully, the driver will need to use some cognition, including memory or imagery, as well as conflict monitoring, response selection and strategic processing. To the best of our knowledge, although many studies have investigated the neural network of top-down control,8,10,11 no functional neuroimaging studies have been performed with the objective of investigating the neural network of response selection and top-down control on the basis of experience or imagery.
Nor has the component of this network which is responsible for the use of the previously acquired information been elucidated. To identify this neural network, in the present study, we developed a modified Stroop task, consisting of a cue followed by a color word Stroop test. To perform the Stroop interference task correctly in this experiment, the subject was required to keep the previously presented cue in his or her memory and retrieve it. While performing this task, continuous attention, imagery or memory were necessary. Using functional magnetic resonance imaging (fMRI) with our modified Stroop task, we attempted to determine which brain regions form a spatially distributed network of top-down control using previously acquired information, and which component of this network is responsible for the use of the previously acquired information.
Methods and Materials
Subjects
Ten healthy volunteers (2 men and 8 women, age=30.3 6.3 years, full-time education=13.3 1.3 years) with no history of neurological disease, psychiatric disease, or drug or alcohol abuse were recruited. All of the subjects were strongly right-handed, as assessed by the modified Edinburgh Handedness Inventory,12 had normal or corrected-to-normal vision and were native Korean speakers. The Ethical Committee approved the protocol, and informed written consent to participate in the present study was obtained from all subjects.
Modified Stroop Task
Each single trial comprised a cue, delay period, probe, and intertrial interval (ITI). A red, blue or yellow colored diagram of a T-shirt was presented as a cue, followed 4 seconds later by a colored word as a probe. Both the cue and the probe were presented for 0.5 seconds. The ITIs, i.e. the durations from the previous probe to the next cue, were 4, 8 or 12 seconds. The subjects had to respond by pressing a button, in order to indicate whether the color of the word was consistent with that of the T-shirt presented 4 seconds earlier. During the 5-minute duration of the modified Stroop task, 30 trials, consisting of 15 facilitation and 15 interference probes, were presented pseudorandomly. Each trial involved 8 YES-answers and 7 NO-answers. The stimuli were presented to the subjects using an LCD projector, back-projected onto a screen placed at the subjects' feet. Motor responses were made using a two-button mouse and were recorded by stimulus presentation software developed by the authors.
Image Acquisition and Preprocessing
Images were obtained with a General Electric Signa 1.5-T high-speed imaging system (modified by Advanced NMR Systems, Wilmington, MA). One hundred and twenty whole brain images of the blood oxygen level-dependent signal intensity were collected during the performance of the modified Stroop task using a T2*-weighted echoplanar imaging (EPI) sequence (24 axial slices, 5 mm thickness, 3.125 x 3.125 mm in-plan resolution, TR=2500 ms, TE=50 ms, flip angle=90º). Image preprocessing and voxelwise analyses were conducted using SPM2 (Statistical Parametric Mapping program developed by the Wellcome Department of Cognitive Neurology, London, UK) software program. The EPI data were normalized to a common mean, movement-corrected using a six-parameter rigid body translation and.68 smoothed using a three-dimensional Gaussian filter (8- mm full width at half maximum) to accommodate between-subject differences in the brain anatomy.
Statistical analysis
The fMRI model was established using the General Linear model with a design matrix comprising 4 derivatives for 1 cue followed by a facilitatory probe (Cf), 1 cue followed by an interference probe (Ci), 1 facilitation (F), and 1 interference (I) conditions. The contrasts, [I-F] and [(I-Ci)-(F-Cf)], were used to identify the networks used to solve the Stroop interference test with previously acquired information and the Stroop interference test only, respectively. To make inferences at a single-group level, these images were analyzed with one-sample t-tests on a voxel-by-voxel basis. To find the neural correlates of the retrieval of the previously acquired information that was used to solve the Stroop interference test, the I-F contrast mask was used. A significance level of
P <.001 (uncorrected for multiple comparisons) and a three-dimensional contiguity of 8-voxels, on the basis of the study of Forman and colleagues,13 were used as the thresholds for the statistical maps. The anatomical localization of suprathreshold activity was determined by overlaying the activation maps onto the reference structural image and by transforming them into standard reporting coordinates using SPM2.
Results
More extensive regions were needed to solve the Stroop interference test with previously acquired information than for the Stroop-only condition (Table 1). Some regions, the right middle frontal gyrus, right precuneus, left cerebellum, right supplement motor area, left insula, and left middle temporal and occipital gyri, were activated to solve the Stroop interference test with previously acquired information. Others, the left ACC, right inferior occipital gyrus, and left postcentral gyrus, were activated to solve the Stroop-only condition.
After using the Stroop, the I-F contrast, mask, the left precuneus (z=3.47; voxel size=10; uncorrected
P <.001) and right superior temporal gyrus (STG; z=3.43; voxel size=10; uncorrected
P <.001) were remained among regions that were activated to solve the Stroop interference test with previously acquired information (Figure 1).
Discussion
The present study used the modified Stroop condition to explore the neural correlates of the retrieval of previously acquired information during top-down control. More extensive regions were activated to solve the Stroop task with previously acquired information than for the Stroop-only condition. The left precuneus and right STG were related to the retrieval of the previously acquired information that was used to solve the Stroop task.
The activation of the left precuneus appeared to be preferentially motivated by an imagined, rather than a visually perceived, context.14,15,16 The STG is a paralimbic association cortex area involved in multimodal object processing and the analysis of speech-like sounds.17 The role of the STG, which is not the primary auditory cortex but was activated in the present study, is not obvious, but may be related to auditory verbal imagery.18 A previous study reported that the precuneus and superior temporal sulcus (STS) were activated by the visual imagery of famous faces.19 The authors of this face study interpreted this to mean that the STS mediates the processing of information relevant to social communication, such as the eye gaze direction and facial expression. However, its activation may be also related to the naming of famous faces. Therefore, our results suggest that the previously acquired information in the present study, for example, the red T-shirt, was divided and stored with verbal and visual components and then retrieved by the visual imagery of the precuneus and the verbal imagery of the STG, in order to solve the Stroop interference test. We could not determine the core region used for the division and imagery of information. However, a previous connectivity study showed that the top-down mechanism arising in the prefrontal cortex was related to visual imagery.20
The present study demonstrated that more extensive regions were activated to solve the Stroop interference test with previously acquired information than for the Stroop-only condition. Although the association of the ACC with interference is well known,5,7 the activation in the left ACC, left postcentral gyrus and right inferior occipital gyrus was not observed in the Stroop condition with previously acquired information (Table 1). In the present study, however, the activations of both the white matter (x=-16, y=32, z=12, 68 voxels, z score=4.11, uncorrected P <.001: not reported in the results) near the left ACC and of the right ACC itself (x=2, y=18, z=28, 3 voxels, Z score=2.68, uncorrected P <.005: not reported in the results) were observed in the Stroop condition with previously acquired information. The postcentral gyrus, which was activated in the Stroop-only condition, was also activated, but less significantly (left: x=-32, y=-44, z=70, 7 voxels, Z score=2.78; right: x=34, y=-26, z=44, 43 voxels, Z score=3.40, uncorrected
P <.005: not reported in the results), in the Stroop condition with previously acquired information.
It is unclear what caused the difference in the activated loci and signal changes in the ACC and postcentral gyrus between the two conditions, and why the activation of the right inferior occipital gyrus was not observed in the Stroop condition with previously acquired information in the present study. One possible interpretation is that, in the case of top-down control with previously acquired information, areas of the prefrontal cortex, such as the right middle frontal gyrus, as well as the visual imagery related network, may play a more active role than the ACC.
One of the limitations of the present study was our complicated paradigm. In our modified Stroop task, the T-shirt stimuli used as a cue and the color words used as a probe had both visual and verbal components, but were different types of stimuli, namely objects and words, respectively. It is possible that it is this difference that causes the two areas, the left precuneus and right STG, to be activated. However, the right STG, which was activated in the present study, may be a verbal imagery related region,21 because of its location in the vicinity of the STS rather than the primary auditory cortex. The study limitations of a relatively small sample size without any intelligence variable were not considered to be critical.
Conclusion
In the present study, we demonstrated that both verbal and visual imagery were related to the retrieval of the information required to exert top-down control. Further studies are needed using connectivity analysis techniques, such as dynamic causal modeling, in order to clarify the top-down control mechanism used for the storage and imagery of verbal and visual information.
Banich MT, Milham MP, Jacobson BL, Webb A, Wszalek T, Cohen NJ, Kramer AF. Attentional selection and the processing of task-irrelevant information: insights from fMRI examinations of the Stroop task. Prog Brain Res 2001; 134: 459-470.
Cohen JD, Dunbar K, McClelland JL. On the control of automatic processes. a parallel distributed processing account of the Stroop effect. Psychol Rev 1990; 97: 332-361.
Miller EK, Cohen JD. An integrative theory of prefrontal cortex function. Annu Rev Neurosci 2001; 24: 167-202.
Kerns JG, Cohen JD, MacDonald AW 3rd, Cho RY, Stenger VA, Carter CS. Anterior cingulate conflict monitor-ing and adjustments in control. Science 2004; 303: 1023- 1026.
Botvinick MM, Braver TS, Barch DM, Carter CS, Cohen JD. Conflict monitoring and cognitive control. Psychol Rev 2001; 108: 624-652.
Carter CS, Braver TS, Barch DM, Botvinick MM, Noll D, Cohen JD. Anterior cingulate cortex, error detection, and the online monitoring of performance. Science 1998; 280: 747-749.
Barch DM, Braver TS, Sabb FW, Noll DC. Anterior cingu-late and the monitoring of response conflict: evidence from an fMRI study of overt verb generation. J Cogn Neurosci 2000; 12: 298-309.
MacDonald AW, Cohen JD, Stenger VA, Carter CS. Dissociating the role of the dorsolateral prefrontal and anterior cingulate cortex in cognitive control. Science 2000; 288: 1835-1838.
Casey BJ, Thomas KM, Welsh TF, Badgaiyan RD, Eccard CH, Jennings JR, Crone EA. Dissociation of response conflict, attentional selection, and expectancy with functional magnetic resonance imaging. Proc Natl Acad Sci USA 2000; 97: 8728-8733.
Carter CS, Macdonald AM, Botvinick M, Ross LL, Stenger VA, Noll D, Cohen JD. Parsing executive processes: strategic vs. evaluative functions of the anterior cingulate cortex. Proc Natl Acad Sci USA 2000; 97: 1944-1948.
Miller BT, D'Esposito M. Searching for "the Top" in Top-Down Control. Neuron 2005; 48: 535-538.
Oldfield RC. The assessment and analysis of handedness: the Edinburgh inventory. Neuropsychologia 1971; 9: 97-113.
Forman SD, Cohen JD, Fitzgerald M, Eddy WF, Mintun MA, Noll DC. Improved assessment of significant activation in functional magnetic resonance imaging (fMRI): use of a cluster-size threshold. Magn Reson Med 1995; 33: 636-647.
Fletcher PC, Happe F, Frith U, Baker SC, Dolan RJ, Frackowiak RS, Frith CD. Other minds in the brain: a func- tional imaging study of 'theory of mind' in story compre-hension. Cognition 1995; 57: 109-128.
Ishai A, Ungerleider LG, Haxby JV. Distributed neural systems for the generation of visual images. Neuron 2000; 28: 979-990.
Mellet E, Tzourio-Mazoyer N, Bricogne S, Mazoyer B, Kosslyn SM, Denis M. Functional anatomy of high-resolution visual mental imagery. J Cogn Neurosci 2000; 12: 98-109.
Jancke L, Wustenberg T, Scheich H, Heinze HJ. Phonetic perception and the temporal cortex. Neuroimage 2002; 15: 733-746.
Jancke L, Shah NJ. Hearing syllables by seeing visual stimuli. Eur J Neurosci 2004; 19: 2603-2608.
Ishai A, Haxby JV, Ungerleider LG. Visual imagery of famous faces: effects of memory and attention revealed by fMRI. Neuroimage 2002; 17: 1729-1741.
Mechelli A, Price CJ, Friston KJ, Ishai A. Where bottom-up meets top-down: neuronal interactions during perception and imagery. Cereb Cortex 2004; 14: 1256-1265.
Aleman A, Formisano E, Koppenhagen H, Hagoort P, de Haan EH, Kahn RS.
The functional neuroanatomy of metri-cal stress evaluation of perceived and imagined spoken words. Cereb Cortex 2005; 15: 221-228.