Serial organization of human behavior in the inferior parietal cortex

Abstract

The parietal cortex is involved in a wide range of cognitive functions in humans including associative functions between multiple sensorimotor spaces, attentional control, and working memory. Little is known, however, about the role and the functional organization of the parietal cortex in action planning and sequential cognition. Moreover, the respective contributions of parietal and frontal regions to action planning remains poorly understood. To address this issue, we designed a functional magnetic resonance imaging protocol requiring subjects to perform overlearned sequences of motor acts and sequences of cognitive tasks. The results reveal only a single bilateral region in the cerebral cortex located in the intraparietal sulcus (IPS; Brodmann's area 40) exhibiting sustained activations during the execution of both motor and task sequences. Additional analyses of phasic activations during sequence execution further suggest a functional dissociation between the left IPS, involved in representing and processing the abstract serial structure of ongoing behavioral sequences regardless of their hierarchical structure, and the right IPS, involved in preparing successive sensorimotor sets that compose such behavioral sequences. We show that this parietal system functionally differs from the frontal system that was previously identified as controlling action selection with respect to the hierarchical rather than serial structure of behavioral plans. Thus, our results reveal the central role of the bilateral intraparietal sulcus in high-order sequential cognition and suggest a major functional segregation within the frontoparietal network mediating action planning, with the frontal and parietal sector involved in processing the hierarchical and serial structure of action plans, respectively.

Figure 1.

Experimental protocol. A, Truncated series of successively presented visual stimuli are shown in both the motor and task condition. Green and red cues were start and stop signals, respectively. In the motor condition, stimuli were square symbols. In each trial, subjects had to select left (L), right (R), or simultaneous left and right (LR) finger presses (as indicated by arrows). In the task condition, stimuli were pseudorandomly chosen letters A, B, or C. In each trial, subjects had to select the sensorimotor task set (TS_A, TS_B, or TS_C as indicated by arrows) to produce the correct motor response to stimuli (indicated below each expected task sets). Examples of endogenous terminations are shown. B, Serial organization of behavioral events in both conditions. C, Hierarchical organization of behavioral sequences in the motor and task condition. See Material and Methods for terminology and additional details.

Figure 2.

Topography of brain activations. Purple indicates regions exhibiting sustained activations during motor and task sequences. Blue indicates regions exhibiting sustained activations during motor sequences only. Green and yellow indicate regions showing all phasic effects [i.e., effects of motor and task increment, motor and task sequence boundary (green, larger effect of motor increment; yellow, larger effect of motor sequence boundaries)]. All these phasic effects were also observed in the left purple region. Orange indicates regions exhibiting effects of motor increment, motor sequence boundary, and task increment but no effect of task sequence boundary. Joint effects of motor sequence boundary and task increment were also observed in the blue and right purple regions. Red indicates regions exhibiting only effects of initiation and endogenous termination in motor and task sequences. Activations are superimposed on anatomical axial slices averaged across subjects (neurological convention) and indexed by the vertical Talairach coordinate (z). Frontal activations are not shown here [reported by Koechlin and Jubault (2006)]. Talairach coordinates of activation peaks are provided in Tables 1 and 2. L, Left; R, right.

Figure 3.

Activations of parietal areas showing both sustained and phasic responses. Graphs show reconstruction of magnetic resonance (MR) signal changes in the different conditions in specified regions averaged over clusters and subjects (error bars indicate SEs across subjects). For each time course, data points are adjusted; peristimulus MR signals are averaged in time bins of 2 s and obtained after subtracting the estimated contribution of other events computed from the multiple linear regression model. A, Sustained effects in the left middle IPS, right dorsal IPS (purple regions in Fig. 2), and right ventral IPS (blue region). The orange line indicates epoch-related activations during task sequence execution. The blue line indicates epoch-related activations during motor sequence execution. Green and red bars indicate when start and stop signals occurred. The late onsets of sustained activations result from the substraction of phasic responses associated with sequence initiation. B, Phasic, event-related activations in the same regions. Dashed lines indicate baseline trials. Solid lines with no symbols indicate increment trials. Solid lines with squares indicate sequence boundary trials. x-axis origins are stimulus onsets.

Figure 4.

Activations of parietal regions showing phasic responses only. Graphs show reconstruction of magnetic resonance (MR) signal changes in the different conditions in specified regions averaged over clusters and subjects (error bars indicate SEs across subjects). For each time course, data points are adjusted; peristimulus MR signals are averaged in time bins of 2 s and obtained after subtracting the estimated contribution of other events computed from the multiple linear regression model. Event-related activations in the left anterior and posterior IPS (yellow and green regions in Fig. 2, respectively), right middle IPS and the precuneus (orange regions), and in bilateral SMG/TPJ regions (red regions) are shown. Dashed lines indicate baseline trials. Solid lines with no symbols indicate increment trials. Solid lines with squares indicate sequence boundary trials. x-axis origins are stimulus onsets. SMG/TPJ activations actually differed between endogenous and exogenous terminations (see Results and Fig. 5)

Figure 5.

Phasic activations in the SMG/TPJ. Graphs show reconstruction of magnetic resonance (MR) signal changes in the different conditions in the SMG/TPJ averaged over clusters and subjects (error bars indicate SEs across subjects). For each time course, data points are adjusted; peristimulus MR signals are averaged in time bins of 2 s and obtained after subtracting the estimated contribution of other events computed from the multiple linear regression model. x-axis origins are stimulus onsets. Triangles indicate responses to sequence initiation. Dashed lines indicate responses in baseline trials. Solid lines indicate responses to sequence increment. Open and filled circles indicate responses to exogenous and endogenous termination, respectively.

Figure 6.

Schematic diagram of parietal and frontal activations observed in the protocol. Parietal activations are represented as solid circles and rectangles. Green and yellow colors indicate the interaction between motor increments and sequence boundaries effects in the anterior and posterior left IPS (green, larger effect of motor increment; yellow, larger effect of motor sequence boundaries). Frontal activations found in the same experimental protocol and reported by Koechlin and Jubault (2006) are shown as open circles and rectangles. Pcu, Precuneus; pSMA, presupplementary motor area.