Bruin Research Bullefin, Vol. 16. pp. 15-17, 1986. o Ankho International Inc. F’rintd in the U.S.A. 0361~923&M $3.00 + .OO

Role of Bar Press-Related Neurons in the Dorsolaterall Pretiontal Cortex During Task Performance by Monkey

MASUMI INOUE,’ YUTAKA OOMURA, HIT00 NISHINO* AND SUJIT KUMAR SIKDAR

Department of Biological Control, National Institute for Physiological Sciences, Okazaki 444 Department of Physiology, Faculty of Medicine, Kyushu University 60

Fukuoka 812 and *Department of Physioiogy, Faculty of Medicine

Toyama Medical and Pharmaceutical University, Toyama 930-01, Japan

Received 28 October 1985

INOUE, M., Y. OOMURA, H. NISHINO AND S. K. SIKDAR. Role of bar press-related neurons in the dorsolateral

prefrontal cortex during task performance by monkey. BRAIN RES BULL 16(l) 15-17, I!%.-Extracellular single neuron activity of the dorsolaterai prefrontal cortex (DL) was recorded in the monkey, during bar pressing for reward. The bar press-related neurons which exhibited excitation or inhibition during the bar press period were found lo be scattered diffusely in the DL. Activity changes that arose during the bar press period also appea& when the experimenter pressed the bar for the monkey. When delivery of food was delayed for a random time after cue tone on, bar press responses were still confined to the bar press period and did not extend beyond the cue tone. These results, together with the lesion studies, suggest that bar press-related neurons are involved in the animal’s concentration during the bar press period.

Monkey Dorsolateral prefrontal cortex Bar press-related neuron Concentration

EVER since Fuster [4] reported single cell recordings from the dorsolateral prefrontal cortex (DL) during delayed task performance, many electrophysiological studies [5, 10, 11, 131 have been carried out to elucidate the functions of this area. These studies indicated that the DL may function in spatial memory and preparation for movement. Deficits in these functions can explain some of the disorders exhibited by monkeys after a lesion of DL, i.e., impairment in delayed responses or delayed alternation tasks [7j. However, it has been possible to compensate for impairments seen in these tasks by forcing the animal’s attention to the task [3]. In addition, reversible depression of prefrontal cortex activity by cooling renders delayed task performance more vulnera- ble to distraction than observed in the normal animal [6]. In other words, the monkey’s capacity to concentrate on a task diminishes with DL dysfunction.

Recently, Ono et al. [IS] reported that a monkey with a large ablation of the DL displayed impaired performance in bar pressing at a high fued ratio for a reward, even though it could successfully carry out bar pressing at a small fured ratio. Their results implied that an ablated monkey could not concentrate consistently on bar pressing for a reward in the presence of interference by various stimuli. In the present experiment, we used a similar bar press task and studied DL neuronal responses at the single cell level to obtain further insight into the DL role in task performance.

Two macaque monkeys (Macaca mulatta and Maraca fuscara) were trained to discriminate between cue lamp on or off during a bar press task. Figure IB shows the sequence of the bar press task: (1) a cue lamp on to signal the start of a trial; (2) a fixed ratio bar press task (usually FR 15); (3) an auditory cue to signal presentation of a reward with a 0.5-I .5 set delay; (4) delivery of reward (a 5 mm baH of bread); (5) taking and eating reward. The experimental procedures have been described elsewhere [8]. Each neuron was charac- terized by its firing pattern during the bar press task. The fling pattern was evaluated with peristimulus time histo- grams (time 0: cue tone on) for 5-15 trials.

Activities from a total of 266 neurons in the DL were recorded during bar press tasks. Of these, 140 neurons re- sponded during either the bar press or the gaze period or both, i.e., from the time of cue lamp on to the time of putting the reward into the mouth. Responses were classified into three groups according to the fling patterns shown in Fig. IC. The type I response (69/M& 4%) were either inhibition (n=27,4W) (Fig. lC, Ia) or excitation (n=42,60%) (Fig. IC, Ib) during the bar press period. Some of these neurons changed their activity transiently at the cue lamp on, cue tone on and/or while eating food as a reward. Activity of type II neurons (33/14O, 24%) was either lower (28/33, 85%) (Fig. lC, IIa)orhigher(5/33,l%)(Fig. lC.IIb)thanbaseline activity while gazing at the food box. Some of these gaze-

‘Requests for reprints should be addressed to Y. Oomura. M. D.. Department of Physiology. Faculty of Medicine, Kyushu University 60, Fukuoka 812. Japan.

B

Ear - cue n

kinto mouth Cl3P-m

Gaze

a b

FIG. 1. A: schematic illustration of experimental set up. The mon- key faced a bar press panel containing a food bar, a cue lamp and a food box. B: sequence of the bar press task. C: examples of typical responses. Type I, activity changes during bar press period. Type II. responses during gaze period. Type III, continuous response from beginning of bar press tilI the time of putting food into mouth. a and b, inhibitory and exitatory response, respectively. Each, IO to I5 trials.

related neurons also responded transiently at other times, such as cue lamp on. These type seemed to have charac- teristics similar to those of the gaze neurons described by Suzuki and Azuma [16]. The activity of 34 neurons of type III (34/140, 24%) changed during the bar press period and persisted until the gaze period (Fig. IC, IIIa, IIIb). The firing rate of most of these neurons (28/34, 82%) was found to be inhibited. Four neurons (4/140, 3%) exhibited excitation dur- ing the bar press period, which changed into inhibition dur- ing the gaze period. The reverse pattern, i.e., inhibition dur- ing the bar press followed by excitation during the gaze period, was not observed.

Two modifications of the task were investigated to de- termine whether or not the neuronal bar press response was coupled to the animal’s bar press behavior. Neuronal activ- ity was recorded in trials in which the experimenter pressed the bar for the animal while the other task parameters were maintained. In the first few trials of this modified task, the monkey searched for the bar when the cue lamp went on, but he soon became familiar with this new paradigm and usually gazed at the food box whenever the experimenter began to press the bar. The magnitude of neuronal bar press re- sponses reflected the learning process during this modified task. In the first several trials, in which the monkey was not familiar with the procedure, the bar press response was less apparent than in later trials. Figure 2B shows the inhibitory bar press response during the first 10 trials of this modified task. Figure 2A is the control, i.e., the normal bar press task. The same results were obtained for all 5 other neurons tested (total, 6), and for all 5 excitatory bar press responses that were tested in the same way. These results can raise the possibility that the neuronal bar press response might be due to some factor other than the physical act of bar pressing such as arousal or motivation. If this were true, the response would not be restricted only to the bar press period, but would also extend to the consumatory period after cue tone

impulse

V-dC’; f”d% ,J++),,w EOG

FIG. 2. Bar press response under various experimental conditions. A: standard bar press task. B: task without bar press by monkey, but the experimenter pressed bar for monkey. Other schedules, same as in standard task. C: task with delayed presentation of rewvard. Deliv- ery of reward, postponed randomly after cue tone on. Other schedules, same as in standard task. It is evident in the elec- trooculogram (EOG) trace that monkey gazed at the food box for the delayed period. Upper, histogram, 10 trials: bar press; cue lamp on and off, and food into mouth: EOG (time constant, 2 set).

on, like type III. Therefore, type I neurons seem to reflect the animal’s concentration during the bar press period whether the animal or the experimenter presses the bar. Fig- ure 2C, which shows response for a delayed reward, verifies this assumption. The bar press response was restricted to the bar press period even though presentation of food was ran- domly delayed. A total of 6 neurons of type I (3 each in type Ia and Ib) were tested for response changes when reward was randomly delayed. All these responses were similar to each other as shown in Fig. 2C.

Lesion studies [IS] have shown that removal of a major part of the DL produces deficit in the performance of high fixed ratio bar press tasks, while low fixed ratio bar press performance is not impaired. These behavioral studies may imply that the lesioned monkey lack the neuronal apparatus required for the normal capacity to concentrate by suppres- sing various forms of external interference. As mentioned above, the bar press response did not reflect arousal or moti- vation. In addition, this response was neither coupled to the animal’s execution of the task, since changes in activity were elicited whether the animal or the experimenter pressed the bar. These findings suggest the possibility that bar press- related neurons are involved in the neuronal apparatus that controls the degree of concentration during the bar press period.

In general, information of the DL is transmitted to the motor cortex through two different pathways [ 121; the first, via the premotor cortex [l]; and the second, through the caudate nucleus [9]. However, what kind of information is transmitted to the motor cortex through these two different routes as yet remains to be elucidated. Bar press-related neurons have also been demonstrated in the head of the cau- date nucleus [14], especially in the dorsolateral part, which receives fiber projection from the DL [9]. Also, cooling of the DL depressed the increased tiring of caudate neurons seen during the bar press period, not in other periods [14].

ROLE OF BAR PRESS-RELATED NEURONS 17

These findings suggest that the bar press-related neurons in the DL drive the caudate bar press-related neurons and this fronto-caudate connection plays an important role in the per-

fomance of high fured ratio bar press task as well as other task performances [2].

ACKNOWLEDGEMENTS

We thank Professors A. Simpson and K. P. Puthuraya for their help in preparation of this manuscript. Thanks are also due to Dr. Aou for critical reading of the manuscript. This work was partly supported by Grants-in Aid for Scientific Research (Y.0) 244021, 444103 and 587035 from the Ministry of Education, Science and Culture.

REFERENCES

1. Arikuni, T., M. Sal&, I. Hamada and K. Kubota. Topo- graphical projections from the prefrontal cortex to the post- arcuate area in the rhesus monkey, studied by retrograde axonal transport of horseradish peroxidase. Neurosci Lert 19: 155-160, 1980.

2. Battig, K., H. E. Rosvold and M. Mishkin. Comparison of the effects of frontal and caudate lesions on delayed response and alternation in monkeys. J Camp Physiol Psycho1 53: 40@404, 1960.

3. Blake, M. O., D. R. Meyer and P. M. Meyer. Enforced obser- vation in delayed response learning by frontal monkeys. J Comp Physiol Psycho1 61: 374-379, 1966.

4. Fuster, J. M. and G. E. Alexander. Neuron activity related to short-term memory. Science 173: 652654, 1971.

5. Fuster, J. M. Unit activity in prefrontal cortex during delayed- - _ response performance: neuronal correlates of transient mem- ory. J Ncurophvsiul36: 61-78. 1973.

6. Fuster, J. M: and R. H. Baueri Visual short-term memory defi- cit from hypothermia of frontal cortex. Brain Res 81: 393-400, 1974.

7. Fuster, J. M. Prefrontal cortex in motor control. In: Handbook of Physiology. The Nervous System Vol. II, edited by V. B. Brooks. Baltimore: Williams and Wilkins, 1981, pp. 11491178.

8. Inoue, M., Y. Oomura, H. Nishino, S. Aou, S. K. Sikdar, M. Hynes, Y. Mizuno and T. Katabuchi. Cholinergic role in mon- key dorsolateral prefrontal cortex during bar-press feeding be- havior. Brain Res 278: 185-194, 1983.

. ._. 9. Kern, J. M. and T. P. S. Powell. The co&o-striate projection in

the monkey. Brain 93~ 525-546. 1970. 10. Kubota, K. and H. Niki. Prefrontal cortical unit activity and

delayed alternation performance in monkeys. J Neurophysiol 34: 337-347, 1971.

11. Kubota, K., T. Iwamoto and H. Suzuki. Visuokinetic activities of primate prefrontal neurons during delayed-response perform- ance. J Neurophysiol37: 1197-1212, 1974.

12. Kubota, K. An introduction to “voluntary movement and the brain.” Adv Neural Sci 28: 3-6, 1984. (In Japanese).

13. Niki, H. Differential activity of prefrontal units during right and left delayed response trials. Brain Res 70: 346-349, 1974.

14. Nishino, H., T. Ono, K. Sasaki, M. Fukuda and K-I. Muramoto. Caudate unit activity during operant feeding behav- ior in monkeys and modulation by cooling prefrontal cortex. Behav Brain Res .ll: 21-33, 1984.

15. Ono, T.,’ H. Nishino, K. Sasaki, M. Fukuda, K-I Muramoto and N. Niida. Functional significance of the lateral hypothala- mus and the prefrontal cortex in feeding behavior. Neurosci Left Suppl 9: S 63, 1982.

16. Suzuki, H. and M. Axuma. Prefrontal neuronal activity during gazing at a light spot in the monkey. Brain Res 126: 497-508, 1977.