SHORTCOMMUNICATION Context-related representation of timing processes in monkey motor cortex Se Âbastien Roux, Miche Ále Coulmance  and Alexa Riehle Institut de Neurosciences Physiologiques et Cognitives, INPC-CNRS, 31 chemin Joseph Aiguier, 13402 Marseille Cedex 20, France Keywords: pointing movements, response probability, single-neuron recording, spatial and temporal uncertainty, time estimation Abstract Precise timing is essential for motor performance, though the neuronal representation of time is unknown. To explore neuronal correlatesoftimingprocessesduringmovementpreparationandexecution,wecomparedtheactivitiesofneuronsrecordedinmonkey motor cortex during the performance of two tasks requiring correct time estimation in different contextual situations. Despite the fact thatbothtaskshadmanycommonfeatures(e.g.samemovements,sametimeintervalstoestimate),theyinvolveddifferentdegreesof spatial and temporal uncertainty. Our ®ndings indicate that time is indeed represented in neuronal activity of the motor cortex, albeit strongly dependent on context and not in an invariant manner as a distinct process per se. Introduction The accurate estimation of time intervals is an essential aspect of motor performance. Most often, temporal judgements are tightly linked to the dynamics of spatial features in the environment, e.g. the time for a ball to arrive within reach. It is, however, not clear if time by itself is represented in the central nervous system ± devoid of spatial associa- tion. In contrast, it is well known that prior information about spatial parameters such as direction strongly modulates neuronal activity in many parts of the motor system, long before movement execution (Weinrich & Wise, 1982; Riehle & Requin, 1989; Wise, 1993; Bastian et al., 1998). This modulation in activity is often de®ned in terms of processes, which speci®cally facilitate movement initiation (Tanji & Evarts, 1976; Kubota & Hamada, 1979; Riehle & Requin, 1989, 1993; Requin et al., 1992). Alternatively, it is discussed in terms of antici- pation and/or signal expectancy without explicitly talking about timing (Mauritz & Wise, 1986; Vaadia etal., 1988). Although the initiation of any voluntary movement requires a correct estimation of the moment when to move, the notion of time estimation or even timing is rarely considered (but see Niki & Watanabe, 1979; Leon & Shadlen, 2003). As a matter of fact, in all the above mentioned studies, time estimation was not a necessary constituent of the task because the moment `when to move' was systematically indicated by a signal. This does not exclude, however, that the subject did use temporal informa- tion. Indeed, we have shown previously that, when the delay between the informative preparatory signal and the imperative response signal (the `preparatory period') is chosen at random but with equal prob- ability from a pretrained set of durations, not only behavioural reaction time decreases as time goes on during the trial ± as a function of increasing the probability to respond ±, but also motor cortical neurons synchronize their activity signi®cantly at moments when a response signal is expected (Riehle et al., 1997, 2000). This clearly implies a time estimation process during movement preparation. In order to study the implication of these processes in more detail, we designed two conceptually different tasks in which time estimation is a neces- sary requisite for correct performance. The aim was to determine whether time estimation is represented invariantly by neuronal activity or is context-dependent in relation to task requirements. Preliminary data have been presented in (Riehle, 2001). Experimental procedures Behavioural tasks A male Rhesus monkey (Macaca mulatta) was used for the study and cared for in the manner described in the Guiding Principles in the Care and Use of Animals in the American Physiological Society and French government regulations. It was trained to execute movements in two opposite directions. On a vertical panel, three touch sensitive targets with light emitting diodes (diameter, 3 cm) were mounted in a hor- izontal line, 10cm from each other. Both lateral targets could be lit either in green or in red, whereas the centre target was lit in yellow. The animal had to touch the centre target with the right hand for initiating the trial. After a ®xed delay of 500 ms, a preparatory signal (PS) was presented. Two different tasks were to be performed, presented in blocks with changing order. (i) In the ®rst task, a choice reaction time task (chRT), both peripheral targets were presented simultaneously as PS, one in red and the other in green. An auditory (directionally noninformative) response signal (RS) then followed after either 600 or 1200 ms. The animal had been trained to associate to each colour one of these two possible delay durations. If the RS occurred after 600 ms, the monkey had to select the red target, whereas if it occurred after 1200 ms, the European Journal of Neuroscience, Vol. 18, pp. 1011±1016, 2003 ß Federation of European Neuroscience Societies doi:10.1046/j.1460-9568.2003.02792.x Correspondence: Dr Alexa Riehle, as above. E-mail: ariehle@lnf.cnrs-mrs.fr  Present address: Ecole Polytechnique Fe Âde Ârale de Lausanne, EPFL-SIC CP 121, 1015 Lausanne, Switzerland. Received 20 January 2003, revised 20 May 2003, accepted 29 May 2003