Edited by: Shuhei Yamaguchi, Shimane University, Japan
Reviewed by: Tom Verguts, Ghent University, Belgium; Tilmann A. Klein, Max Planck Institute for Human Cognitive and Brain Sciences, Germany
*Correspondence: Yael Benn, Department of Psychology, University of Sheffield, Western Bank, Sheffield, S10 2TN, UK e-mail:
This article was submitted to the journal Frontiers in Human Neuroscience.
This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) or licensor are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.
The neural basis of progress monitoring has received relatively little attention compared to other sub-processes that are involved in goal directed behavior such as motor control and response inhibition. Studies of error-monitoring have identified the dorsal anterior cingulate cortex (dACC) as a structure that is sensitive to conflict detection, and triggers corrective action. However, monitoring goal progress involves monitoring correct as well as erroneous events over a period of time. In the present research, 20 healthy participants underwent functional magnetic resonance imagining (fMRI) while playing a game that involved monitoring progress toward either a numerical or a visuo-spatial target. The findings confirmed the role of the dACC in detecting situations in which the current state may conflict with the desired state, but also revealed activations in the frontal and parietal regions, pointing to the involvement of processes such as attention and working memory (WM) in monitoring progress over time. In addition, activation of the cuneus was associated with monitoring progress toward a specific target presented in the visual modality. This is the first time that activation in this region has been linked to higher-order processing of goal-relevant information, rather than low-level anticipation of visual stimuli. Taken together, these findings identify the neural substrates involved in monitoring progress over time, and how these extend beyond activations observed in conflict and error monitoring.
The majority of human activity is goal-directed (Locke,
Progress monitoring, however, involves more than simply the detection of errors, and reviews (e.g., Carver and Scheier,
A range of evidence points to the likely neural basis of processes involved in progress monitoring (for a review, see Berkman and Lieberman,
Attention is likely to be the first process that is engaged when monitoring goal progress. Attention is defined as the process that selects which sensory information is processed (and possibly reaches awareness) at any one time. As such, attention is considered to be a multi-channel process that can be driven by either bottom–up (e.g., when attention is diverted to a salient stimulus) or top–down (e.g., goal directed) processes. While stimulus-driven attention is mostly supported by the ventral network (that consists of the lateral and inferior frontal/prefrontal cortex (PFC) and the temporo-parietal junction, Corbetta and Shulman,
How bottom–up and top–down systems interact to control attention has been the focus of considerable research in recent years (e.g., Buschman and Miller,
Finally, there is some evidence that top–down directed attention can modulate activity in early sensory cortices, resulting in regions such as the primary visual cortex, that are often associated with processing of existing visual stimuli, being activated in anticipation of a specific visual stimulus (Chawla et al.,
Attending to information is the first step involved in progress monitoring. However, this information then needs to be stored and updated as new information becomes available. Storage and updating is considered to involve WM (Baddeley,
Behaviorally and from a neuroscience perspective, there seem to be close relationships, and possibly an overlap, between processes involved in tasks that engage WM and those that involve selective attention (Gazzaley and Nobre,
Once the current state has been identified and committed to memory, the person seeking to evaluate their goal progress must compare that information to their reference value or goal and identify any discrepancies. Most research to date has investigated the neural basis of discrepancy detection (i.e., recognizing that the current state does not match the reference value) using paradigms involving error detection (i.e., identifying conflict between the current and desired response). For example, the first study using functional magnetic resonance imagining (fMRI) to examine error monitoring (Carter et al.,
Since Carter et al.'s study, similar paradigms have been studied using a range of neuroimaging techniques, including intracranial recordings (Gehring et al.,
Despite a range of evidence on the likely neural regions involved in monitoring goal progress, several conceptual and methodological issues remain unanswered. Monitoring goal progress is a complex process that simultaneously involves elements or combinations of the processes described above. That is to say, while these sub-processes are temporally organized, they are not easily distinguishable (e.g., WM and attention overlap, as discussed above), and processes are not necessarily evoked in a linear temporal fashion. For example, Control Theory (Carver and Scheier,
In an effort to overcome this problem, Berkman et al. (
There are a number of other limitations to studying the neural basis of progress monitoring using tasks that only involve trial-by-trial monitoring. Typically, monitoring goal progress requires that the person periodically attend to relevant information (Berger,
Finally, as noted earlier, previous studies have focused on conflict monitoring, and have tended to employ paradigms involving error detection. However, monitoring goal progress is fundamentally different to monitoring for errors, as progress monitoring is an ongoing process where relevant information has to be updated, or aggregated, in order to assess the current state in relation to the goal. In contrast, error monitoring involves identifying discrete errors. Furthermore, while conflict monitoring focuses on detecting errors, or detecting situations where errors may occur, progress monitoring involves both the detection of erroneous and correct responses (i.e., in both instances people ask themselves “did I get that one right?”). Therefore, focusing on the brain regions activated by erroneous, compared with correct, responses is likely to mask the areas that are involved in progress monitoring.
The present research investigated the neural basis of monitoring progress over a medium term (rather than on a trial-by-trial basis) in a context that does not rely on comparing the neural substrates of erroneous responses to correct ones. Two computer games were designed whose conditions differed in the nature of the progress monitoring that was required. One computer game involved processing numerical stimuli, since progress monitoring often involves numerical processing (e.g., checking finances, counting calories, etc.) and evidence suggests that people find it easier to monitor quantifiable outcomes (Josephs et al.,
Given the role of the parietal cortex in selective attention and WM (Gazzaley and Nobre,
Ethical approval for the study was granted by the Ethics Sub-Committee in the Department of Psychology at The University of Sheffield. The research was conducted in a manner consistent with the American Psychological Association's ethical principles.
Twenty healthy, native English speaking, right-handed (as assessed by the revised Edinburgh handedness questionnaire, Oldfield,
The tasks were developed using Pygame 1.9.1. Participants played two different games: a numerical game termed the “harbormaster game” and a visuo-spatial game termed the “nursery game.” In the harbormaster game, participants were asked to play the role of a harbormaster at a busy port. As such, participants were presented with three boats arriving at the port on each trial, and asked to decide which of the three boats should enter the port (by selecting a boat using an MR-compatible mouse; NAtA technologies, FOM-2B-10B fMRI Mouse). There were three different types of boats indicated by different values; boats marked with a zero represented tourist boats (no fish bought or sold), boats with a positive number represented fishing boats (bringing fish into the port, with the amount of fish indicated by the number on the boat), and boats with a negative number represented merchant boats (wishing to buy the amount of fish indicated by the number on the boat; Figure
There were four conditions in the harbormaster game: In the
In the condition involving
In each of the conditions, after every 4–6 trials, participants were asked a question to test whether they had followed the instructions. In Conditions 3 and 4, participants were asked about the current fish stocks at the port. In Condition 2, they were asked to report the sum of the two non-zero boats that had arrived at the port on the last trial. In Condition 1, participants were given a simple calculation question, similar to the one required in Condition 2, but unrelated to the task (e.g., the sum of −2 and 0, as illustrated in Figure
In the nursery game, participants were asked to imagine themselves as an infant in a nursery being presented with three trucks to play with. On each trial, they were asked to select which of the three trucks appearing on the screen they wished to play with. There were always two trucks marked with a randomly selected shape representing toy blocks, and one truck that was not marked with a shape—the “empty truck” (Figure
The nursery game had two conditions. In the
In both games the vehicles were animated to move toward the participant for 1.5 s before the participant was able to make a selection. Two aspects of participants' performance were measured during the tasks. First, we recorded which vehicle participants selected on each trial. This information was used to verify that participants followed the instructions. Responses were scored by assigning one point every time a participant selected an appropriate vehicle (e.g., selected a tourist boat in the no monitoring condition of the harbormaster game), and zero points if they selected the wrong vehicle (e.g., selected a tourist boat on a trading day). Second, we recorded participants' responses to the questions designed to check that participants were monitoring progress as required in the harbormaster game.
Participants were given instructions for each game and an opportunity to practice all conditions on a computer before entering the scanner. Once lying in the scanner, participants practiced the games a second time to familiarize themselves with the MRI compatible mouse. Stimuli were viewed on a LCD screen via a head-coil-mounted, rear-facing mirror.
The experiment used a block design, with three runs (Figure
All MR images were acquired at 3T (Ingenia 3.0T, Philips Healthcare, Best Holland) using a fifteen-channel radiofrequency receive-only head coil. Cerebral vascular response to the tasks was recorded using the blood oxygenation level-dependent (BOLD) T2*-weighted signal time-course. During each functional scan, a time series of 194 dynamic datasets was obtained using a 2-dimensional single-shot, echo-planar imaging (EPI) sequence. The EPI scan parameters were as follows: repetition time (TR) = 3000 ms; echo time (TE) = 35 ms; sensitivity-encoding factor = 1.8; flip angle = 90°; in-plane voxel size = 2.4 × 2.4 mm interpolated to 1.8 × 1.8 mm; 35 contiguous 2-dimensional transaxial slices each having slice thickness = 4 mm. Anatomical reference data were obtained for each subject using a Magnetization Prepared-Rapid Acquisition Gradient Echo (MP-RAGE) technique (TE = 3.8 ms; TR = 8.3 ms; TI = 963 ms; flip angle = 8°). This 3D-encoded acquisition yielded T1-weighted data covering the entire intra-cranial structures at a voxel resolution of 1 × 1 × 1 mm.
Data analysis was performed using SPM8 (Wellcome Department of Imaging Neuroscience, London;
The mean accuracy with which participants selected the targets (Figure
Given that all conditions had a high rate of correct responses, as reflected by the behavioral data, we did not exclude individual responses from the fMRI analysis as they are unlikely to significantly influence the group data. Furthermore, since a similar level of errors was observed in all conditions, it is likely that activations related to errors would not be statistically significant following subtraction of one condition from another.
There were four conditions in the harbormaster game: no monitoring (Condition 1), trial-by-trial monitoring (Condition 2), monitoring over-time without a reference value (Condition 3), and monitoring progress over-time with respect to a reference value (Condition 4). To examine the neural processes involved in the different types of monitoring, we computed three contrasts: Condition 2 > 1, Condition 3 > 1, and Condition 4 > 1. Contrast 2 > 1 involved right superior parietal, left inferior parietal and bilateral cingulate and superior and medial frontal gyri. Contrast 3 > 1 resulted in activation in the right middle and medial frontal gyri, cingulate gyrus and left inferior and superior parietal regions. Contrast 4 > 1 resulted in activation of the bilateral precuneus, inferior and superior parietal lobules, inferior, medial and middle frontal gyri and cingulate gyrus (Table
Parietal | Superior parietal lobule | – | 30 | −64 | 46 | 4.87 | 1 | 1.03 |
Frontal | Superior (L,C,R)/medial (L,R) frontal gyrus | 6(L,R),8(L) | 0 | 12 | 56 | 5.35 | 19 | 1.38 |
Frontal(L)/limbic(L,R) | Cingulate gyrus | 32 | −8 | 18 | 42 | 5.36 | 27 | 1.16 |
Parietal | Inferior parietal lobule | 40 | −34 | −54 | 48 | 5.25 | 12 | 1.29 |
Inferior parietal lobule | 40 | −44 | −40 | 46 | 5.32 | 25 | 1.54 | |
Frontal | Inferior frontal gyrus | – | −48 | 10 | 20 | 5.0 | 1 | 1.47 |
Frontal | Middle frontal gyrus | 8 | 32 | 22 | 52 | 5.28 | 21 | 2.0 |
Frontal/limbic | Cingulate/medial frontal gyrus | 32 | 4 | 24 | 42 | 6.52 | 213 | 1.91 |
Parietal | Inferior parietal lobule | 40 | 50 | −36 | 48 | 5.39 | 76 | 1.51 |
Parietal | Inferior parietal lobule | 40 | −6 | −74 | 50 | 5.39 | 76 | 1.65 |
Precuneus/superior parietal lobule | 7 | 50 | −36 | 48 | 5.92 | 105 | 1.51 | |
Parietal | Inferior parietal lobule, sub gyral | – | 34 | −60 | 40 | 4.97 | 3 | 1.2 |
Inferior parietal lobule, post-central gyrus | 40 | 50 | −38 | 48 | 5.53 | 97 | 1.35 | |
Inferior parietal lobule | 7 | 36 | −62 | 44 | 4.97 | 2 | 1.33 | |
inferior parietal lobule | 40 | 48 | −60 | 46 | 4.98 | 3 | 1.09 | |
Inferior/superior parietal lobule | 7 | 36 | −68 | 48 | 5.13 | 16 | 1.44 | |
Precuneus, superior parietal lobule | 7 | 8 | −68 | 48 | 5.64 | 77 | 1.68 | |
Frontal | Insula/inferior frontal gyrus/extra nuclear | 47 | 34 | 18 | −6 | 5.57 | 19 | 0.99 |
Middle frontal gyrus | – | −44 | 46 | −4 | 5.26 | 9 | 1.5 | |
Frontal/limbic | Inferior/medial/middle/superior/cingulate frontal gyrus | 6,8,9,46 | 38 | 34 | 32 | 6.37 | 758 | 1.6 |
Limbic | Anterior cingulate | – | 10 | 32 | 26 | 5.01 | 1 | 1.03 |
Cerebellum | Posterior lobe-Uvula | – | −32 | −64 | −34 | 5.0 | 4 | 1.25 |
Parietal | Precuneus, inferior/superior parietal lobule | 7,19,40 | −44 | −40 | 44 | 6.45 | 396 | 1.59 |
Precuneus | 7 | −8 | −72 | 48 | 5.63 | 29 | 1.27 | |
Frontal | Inferior/middle frontal gyrus | 10,46 | −42 | 38 | 18 | 5.17 | 37 | 1.8 |
Middle/-18superior frontal gyrus | 6 | −18 | 14 | 58 | 4.98 | 7 | 1.12 | |
Frontal/limbic | Cingulate/medial frontal gyrus | 8,32 | −2 | 20 | 48 | 5.36 | 22 | 1.76 |
To identify the differences between monitoring over time and trial-by-trial monitoring, we computed the contrast between Conditions 3 > 2, where WM and calculation demands are similar. The results revealed large areas of activation in the right superior and middle frontal gyri (BA8/9/10) (Table
Sub-lobar | Insula/extra nuclear | 13 | 32 | 14 | −6 | 5.14 | 2 | 0.77 |
Extra nuclear | 13 | 30 | 16 | −8 | 4.86 | 1 | 0.65 | |
– | 20 | 14 | 10 | 4.91 | 1 | 0.45 | ||
– | 8 | −4 | 0 | 5.28 | 5 | 0.5 | ||
Frontal | Middle frontal gyrus | 10 | 40 | 50 | 8 | 5.04 | 15 | 1.54 |
Superior frontal gyrus | 10 | 30 | 56 | 20 | 4.91 | 3 | 1.59 | |
Superior/middle frontal gyri | 8/9/10 | 22 | 32 | 42 | 5.97 | 386 | 1.17 | |
Limbic/parietal | Cingulate gyrus/precuneus | 7 | 6 | −36 | 44 | 4.86 | 3 | 0.9 |
Frontal | Middle frontal gyrus | 8 | −24 | 24 | 50 | 4.97 | 1 | 0.83 |
Superior frontal gyrus | 6 | −20 | 12 | 60 | 4.88 | 1 | 0.84 |
To further examine the neural basis of monitoring progress over time with a clear reference value, we computed the contrasts from BOLD imaging data between Conditions 4 > 2 and Conditions 4 > 3 (Table
Parietal | Inferior parietal lobule | 40 | 52 | −44 | 46 | 5.24 | 11 | 1.21 |
Angular gyrus/inferior parietal lobule | 39,40 | 48 | −64 | 38 | 5.04 | 10 | 1.21 | |
Frontal | Middle frontal gyrus | 10,46 | 44 | 44 | 18 | 5.10 | 7 | 1.73 |
– | 26 | 18 | 44 | 4.89 | 1 | 1.19 | ||
Middle/superior frontal gyrus | 9 | 38 | 38 | 34 | 5.81 | 35 | 1.49 | |
Occipital | Cuneus | 17 | −16 | −88 | 4 | 5.37 | 26 | 1.01 |
The nursery game was designed to examine the brain regions involved in monitoring progress toward a visuo-spatial target. To do so, we computed the contrast between Conditions 2 > 1 (Table
Cerebellum | Posterior lobe/Pyramis | – | 24 | −64 | −38 | 5.27 | 7 | 0.90 |
Occipital | Cuneus | 17/28/23/30 | 12 | −74 | 10 | 5.4 | 76 | 1.38 |
18 | 6 | −90 | 18 | 4.99 | 6 | 1.11 | ||
Parietal | Inferior/superior parietal lobule/precuneus/sub gyral | 7/19/39/40 | 32 | −72 | 44 | 6.01 | 301 | 2.21 |
Frontal | Inferior frontal gyrus | 47 | 30 | 24 | −6 | 5.14 | 43 | 2.48 |
Inferior frontal gyrus/sub-gyral | – | 46 | 10 | 18 | 4.95 | 6 | 2.31 | |
Middle frontal gyrus | 10 | 36 | 50 | 6 | 4.97 | 3 | 2.20 | |
9/46 | 48 | 32 | 30 | 5.45 | 100 | 3.17 | ||
Medial frontal gyrus | – | 10 | 36 | 36 | 4.84 | 1 | 1.37 | |
Thalamus | Pulvinar | – | 20 | −30 | 2 | 5.02 | 1 | 0.49 |
– | 22 | −28 | 4 | 4.85 | 1 | 0.48 | ||
Limbic | Anterior cingulate/cingulate gyrus | – | 12 | 26 | 30 | 4.95 | 2 | 1.19 |
Parietal | Precuneus (L,R)/sub-gyral (R)/superior parietal (L) | 7 | 10 | −76 | 46 | 5.82 | 200 | 1.64 |
Limbic | Cingulate gyrus/medial frontal gyrus (L,R) | 6,8,32 (L,R) | 4 | 24 | 46 | 6.17 | 239 | 2.66 |
Cerebellum | Anterior lobe/Culmen | – | −38 | −44 | −26 | 4.84 | 1 | 1.19 |
Occipital | Cuneus | 17 | −12 | −78 | 4 | 5.28 | 33 | 1.29 |
18/19 | −12 | −90 | 22 | 5.18 | 19 | 0.84 | ||
Occipito-temporal | Fusiform gyrus/sub-gyral | 37 | −46 | −60 | −14 | 5.09 | 14 | 2.0 |
Parietal | Inferior/superior parietal lobule/sub gyral | 7/19/40 | −28 | −68 | 42 | 6.08 | 317 | 2.71 |
Inferior parietal lobule | – | −46 | −38 | 44 | 4.98 | 6 | 1.58 | |
Frontal | Insula/inferior frontal gyrus | 13/47 | −30 | 20 | −6 | 5.19 | 21 | 2.0 |
Middle/inferior frontal gyrus/sub-gyral | 46 | −46 | 32 | 22 | 5.66 | 168 | 2.92 | |
Inferior frontal gyrus/sub-gyral | – | −40 | 8 | 24 | 5.11 | 20 | 2.64 | |
Middle frontal gyrus | 6 | −28 | 10 | 62 | 5.71 | 18 | 1.28 | |
Thalamus | Pulvinar | – | −10 | −8 | −2 | 4.96 | 1 | 0.52 |
Sub-lobar | Extra nuclear/corpus callosum | – | −2 | 4 | 24 | 4.86 | 3 | 0.93 |
To identify the neural basis of progress monitoring in a way that is relatively independent of the modality of the target representations (verbal or visual), we computed the combined contrast between Conditions 4 > 1 of the harbormaster game and Conditions 2 > 1 of the nursery game. The results are presented in Table
Occipital | Cuneus | 17,18,23,30 | 14 | −78 | 6 | 5.21 | 73 | 1.32 |
Parietal | Inferior/superior parietal lobule, precuneus, angular gyrus, sub-gyral | 7,19,39,40 | 32 | −70 | 44 | 6.22 | 256 | 2.62 |
Superior parietal lobule, precuneus | 7 | 10 | −78 | 48 | 5.82 | 178 | 2.92 | |
Inferior parietal lobule, post-central gyrus | 40 | 52 | −44 | 50 | 5.83 | 44 | 1.86 | |
Frontal | Inferior frontal gyrus. Insula, sub-gyral | 13,47 | 34 | 20 | −6 | 5.32 | 74 | 2.85 |
Middle/superior fontal gyrus | 10 | 28 | 60 | −8 | 5.31 | 11 | 3.73 | |
Middle frontal gyrus, sub-gyral | 10 | 36 | 50 | 6 | 5.44 | 29 | 3.61 | |
Inferior/middle/superior frontal gyrus | 9,10,46 | 42 | 34 | 34 | 5.71 | 233 | 4.26 | |
Middle/superior frontal gyrus | 6,8 | 30 | 20 | 46 | 6.35 | 188 | 2.43 | |
Limbic/frontal | Anterior cingulate(R), cingulate/medial frontal gyrus(L,R), superior frontal gyrus (L,C,R) | 6,8,32(L,R),9(R) | 6 | 26 | 46 | 6.78 | 530 | 2.66 |
Cerebellum | Posterior lobe-cerebellar tonsil | – | −38 | −58 | −54 | 4.93 | 8 | 1.7 |
Occipital | Cuneus | 18 | −10 | −88 | 12 | 5.09 | 8 | 1.17 |
18,19 | −8 | −94 | 22 | 5.2 | 20 | 1.11 | ||
Parietal | Inferior/superior parietal lobule, precuneus, supramarginal/angular gyrus, sub-gyral | 7,19,39,40 | −28 | −68 | 42 | 6.49 | 588 | 2.72 |
Precuneus, superior parietal lobule | 7 | −6 | −74 | 50 | 6.25 | 119 | 2.17 | |
Frontal | Inferior frontal gyrus/insula/extra nuclear | 13,47 | −28 | 20 | 0 | 5.25 | 29 | 1.73 |
Inferior frontal gyrus, sub-gyral | – | −36 | 38 | 6 | 5.5 | 29 | 1.08 | |
−40 | 10 | 24 | 5.29 | 55 | 2.14 | |||
Middle frontal gyrus | – | −42 | 44 | −6 | 5.52 | 13 | 2.62 | |
6 | −26 | 12 | 46 | 5.01 | 10 | 1.38 | ||
Inferior/middle frontal gyrus, sub-gyral | 46 | −46 | 34 | 22 | 5.82 | 238 | 3.4 | |
Middle/superior/medial frontal gyrus | 6,32 | −24 | 14 | 60 | 5.52 | 76 | 1.63 |
The present research examined the neural regions involved in monitoring goal progress, across numerical and visuo-spatial modalities. Two computer games were designed whose conditions differed in the nature of the progress monitoring that was required. In the harbormaster game, Condition 1 acted as a baseline condition, Condition 2 involved trial-by-trial monitoring, Condition 3 involved monitoring information over time without comparing progress to a reference value, and Condition 4 involved monitoring progress over time with respect to a numerical reference value. In the nursery game, Condition 1 acted as a baseline condition, while Condition 2 involved monitoring progress over time with respect to a reference value of a visual nature. The findings from both games were used to identify a modality-independent network of activations involved in monitoring progress over time.
The findings point to a series of activations in the fronto-parietal network, including the right DLPFC (BA9) and bilateral inferior and superior parietal regions. This network is largely similar to activations identified in studies of the neural basis of attention and WM (Corbetta and Shulman,
To our knowledge, the present research is the first to examine the neural basis of monitoring progress over a medium term, rather than on a trial-by-trial basis. As predicted, all conditions (in comparison to baseline) increased activation in the fronto-parietal network. This included activation of the inferior frontal gyrus, previously shown to be involved in stimulus-driven attention (Corbetta and Shulman,
We hypothesized that the DLPFC would be activated when monitoring goal progress over time, due to its involvement in updating WM (Petrides,
Activations in the inferior and superior parietal cortices were also observed in all conditions involving monitoring (e.g., trial-by-trial and over time) compared with the baseline conditions. It is likely that in some contrasts (namely, contrasts 2 > 1, 3 > 1 and 4 > 1 in the harbormaster game, contrast 2 > 1 in the nursery game) activations of the right superior and left inferior parietal reflect, at least in part, processes involved in calculation (Benn et al.,
In contrast to the consistent activation observed in the fronto-parietal network, the dACC, which has been previously associated with discrepancy detection and reduction (Carter et al.,
As noted earlier, monitoring with respect to a reference value (contrast 4 > 3 of the harbormaster game and the contrast between the conditions of the nursery game) led to activation in early visual processing regions (BA17). This finding is consistent with accumulating evidence that early sensory processing is involved in goal-directed behavior. For example, evidence suggests that top-down directed attention results in increased firing of neurons in the visual cortex (V1 and V4) of rhesus monkeys (McAdams and Maunsell,
Monitoring strategies may differ between individuals. For example, one participant may keep track of the shapes that they have collected by trying to form an aggregated image, while another participant may try to remember individual shapes. However, in the present research we analyzed the findings at the group level, assuming that all participants approached the task in a similar way—which seemed reasonable given the relatively constrained nature of the focal tasks. Future research might, however, wish to examine the neural basis of different monitoring strategies. This could be done in a quasi-experimental fashion by asking participants what strategy (or strategies) they used, or in an experimental fashion by directing participants to use one strategy or another. Relatedly, the behavioral responses in the present experiment (namely, whether participants selected appropriate targets, or could report on the relevant dimension that they were monitoring) were simply designed to ensure that participants followed task instructions, rather than to elucidate the nature or difficulty of monitoring progress. Future research could, however, use behavioral measures to examine the nature of progress monitoring—e.g., examining the effect of different monitoring strategies on performance on a secondary task (e.g., one involving WM). We would expect that more demanding forms of monitoring, such as comparing a current value to a reference value, would impact performance on a secondary task to a greater extent than less demanding forms of monitoring.
In addition, given the role of the parietal cortex in both numerical and visuo-spatial processing and the observed activation in primary visual cortex, one limitation of the present research is that it is difficult to identify the determinants of activations in these regions. Future research could, however, address this issue by using an auditory paradigm. For example, participants could work toward recreating a musical rhythm by selecting sub-patterns of this rhythm. An auditory paradigm could further help to establish whether monitoring goal progress affects activation in other primary sensory cortices, and to what degree parietal activation can be attributed to visuo-spatial, numerical or attentional processing.
The present research investigated the neural correlates of monitoring goal progress over a medium-term period. Our findings largely support the view that the dACC plays a role in discrepancy detection, which is an important aspect of progress monitoring. However, the present research also helped to identify the neural basis of monitoring progress over time, something that has been relatively neglected in previous studies. We found that regions of the parietal cortex, as well as the right DLPFC, are involved in monitoring progress over time—something that is likely due to monitoring placing demands on attention and WM resources as well the requirement for checking and updating of information over time. Lastly, we report evidence that progress monitoring activated regions of the primary visual cortex, adding to growing evidence that the primary sensory cortices may play an important role in monitoring information in relation to specific goals.
The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.
We are grateful to Tom C. Benn, for programming the experimental paradigm. This research was funded by a grant from the European Research Council (ERC-2011-StG-280515).
1On participant was identified whose performance was significantly worse than the average for the group in two out of the six conditions (>2.5 standard deviations from the mean). To see whether inclusion of this participant influenced our findings, we reran the behavioral and fMRI analysis excluding this participant. The results were largely unchanged and so the analyses reported include all participants.