Comparison of neuronal activity in the rostral supplementary and cingulate motor areas during a task with cognitive and motor demands D. Akkal, B. Bioulac, J. Audin and P. Burbaud Laboratoire de Neurophysiologie, UMR CNRS, 5543, Universite  Victor Segalen, 146, rue Le  o Saignat, 33076 Bordeaux, France. Keywords: Macaca mulatta, motor control, rostral medial wall, unit recordings Abstract A number of cortical motor areas have been identi®ed on the medial wall of the hemisphere in monkeys. However, their speci®c role in motor control remains unclear. In this study, we sought to describe and compare the functional properties of the presupplementary (pre-SMA) and rostral cingulate (CMAr) motor areas in two monkeys performing a visually instructed, delayed, sequential movement. We recorded 134 task-related neurons in the pre-SMA and 149 in the CMAr. The main difference between the two areas was the abundance of responses to targets (46%) in the pre-SMA, while CMAr activity was more related to reward (28%). Neuronal responses to targets were more phasic and higher in frequency in the pre-SMA than in the CMAr. During the delay, the percentage of neuronal responses was similar in the two areas. The discharge pattern was different depending upon whether the delay duration was ®xed or variable but in most neurons was the same regardless of the sequence performed. Movement-related changes were common in the pre-SMA (75%) and in the CMAr (81%) but they occurred earlier in the former. Neurons activated exclusively during movement were more numerous in the CMAr. Finally, neuronal activity in the pre-SMA was more related to the sequential aspect of the task compared to the CMAr. Our results suggest that although the two areas share functional properties, they also participate in different aspects of motor behaviour. Their functional properties re¯ect their anatomical positions, which give them the potential to integrate external stimuli (pre-SMA) and internal states (CMAr) during motor planning. Introduction In the last 10 years, there has been an increasing interest in the motor functions of the medial wall (for reviews, see Goldberg, 1985; Dum & Strick, 1993; Tanji, 1994, 1996; Passingham, 1996a, b; Picard & Strick, 1996; Tanji & Shima, 1996a, b). Lesions of this cortical region in humans (Halsband et al., 1993; for review see Devinsky et al., 1995) and in monkeys (Halsband, 1987; Stern & Passingham, 1994; Chen & Wise, 1995) severely affect the planning and initiation of movements. In addition to the supplementary motor area, which was the ®rst medial motor area discovered by Pen®eld & Welch (1951), a number of motor ®elds have been identi®ed on the medial wall of the hemisphere in monkeys (for review, see Picard & Strick, 1996; Dum & Strick, 1991a, b; He et al., 1995). Two of these occupy the rostral portion of the medial wall: the presupplementary motor area (pre- SMA) located in the medial portion of the superior frontal gyrus and the rostral cingulate motor area (CMAr) buried in the cingulate sulcus. A growing number of functional imaging studies in humans have stressed the role of the pre-SMA and CMAr in higher order aspects of motor behaviour (Posner et al., 1988; Deiber et al., 1996, 1999; Hikosaka et al., 1996; Jueptner et al., 1997; Boecker et al., 1998; Petit et al., 1998; Jenkins et al. 2000). In comparison, few single-unit activity studies have explored their contribution to motor control, especially the CMAr, and their role remains unclear. The pre-SMA activity was compared to the activity in the supplementary motor area (SMA) and found to be more related to motor preparation (Matsuzaka et al., 1992), the update of motor plans, or the acquisition of new sequential procedures (Matsuzaka & Tanji, 1996; Shima et al., 1996; Nakamura et al., 1998). The anterior cingulate cortex was compared to the posterior cingulate cortex and seemed to play a preferential role in self-initiated movements (Shima et al., 1991). However, to date, no comparative investigation has been under- taken between the pre-SMA and the CMAr, and it is not known what functional differences may exist between them. Their anatomical organization suggests that they could have different functions. They are interconnected, and both receive a strong and direct input from the prefrontal cortex (Luppino et al., 1990, 1993; Bates & Goldman- Rakic, 1993; Morecraft & Van Hoesen, 1993; Lu et al., 1994). However, only the CMAr has substantial connections with the primary motor cortex and the spinal cord (Morecraft & Van Hoesen, 1992; Luppino et al., 1990, 1993; Dum & Strick, 1991a; He et al., 1995; Morecraft et al., 1997; Wang et al. 2001). Furthermore, the CMAr projects heavily to the SMA (Wang et al. 2001) whereas the pre-SMA connections with SMA are relatively modest (Luppino et al., 1990, 1993). In addition, the CMAr receives stronger inputs from the orbitofrontal cortex (Carmichael & Price, 1995; Cavada et al. 2000) and the limbic system (Baleydier & Mauguiere, 1980; Vogt & Pandya, 1987; Vogt et al., 1987; Morecraft & Van Hoesen, 1993, 1998) than the pre-SMA. Correspondence: Dr Pierre Burbaud, as above. E-mail: Pierre.Burbaud@umr5543.u-bordeaux2.fr Received 27 February 2001, revised 3 January 2002, accepted 16 January 2002 European Journal of Neuroscience, Vol. 15, pp. 887±904, 2002 ã Federation of European Neuroscience Societies