Introduction

Top Abstract Introduction Methods Results Discussion References

Timing behavior is important in many operant schedules, but two reinforcement schedules that are thought to measure timing behavior directly are TRD and the more well known DRL. Under both schedules, the temporal dimension of the response is differentially reinforced, such that reinforcement can be made contingent upon responding which is correlated with a minimum and a maximum time value of that response or the time between responses (Ferraro and Grilly, 1970; McMillan and Patton, 1965; Sidman, 1955, 1956). Under both schedules, the subject presumably relies on interoceptive cues to regulate behavior. The DRL schedule differs from the TRD in the specifics of the behavioral response. Whereas the animal under a DRL is reinforced for spacing responses at a specified time duration, under a TRD the animal must emit a continuous response of specified duration.

The number of drugs that have been studied with subjects performing under TRD schedules is limited. Stimulants, such as MAP, amphetamine, cocaine and caffeine have produced mixed results. Increases in the proportion of responses both too short and too long in duration to produce a reinforcer have produced flattened response-duration distributions in some studies with MAP and cocaine (Hudzik and McMillan, 1994b; McMillan et al., 1994), whereas amphetamine and caffeine have predominantly produced increases in the proportion of shorter response durations (Schulze and Paule, 1990; Buffalo et al., 1993).

The antipsychotic, chlorpromazine, also has produced equivocal results with long-duration responses predominating in some animals and short durations in others within the same study (Ferguson and Paule, 1992). Antidepressants have produced unique profiles under the TRD 1-1.3 sec schedule. Drugs such as imipramine, trazodone and tranylcypromine increase slightly the proportion of both short and long response durations, but the most interesting effect is the production of many long response durations. These long response durations greatly exceed the reinforced response durations causing large increases in the mean response duration at low doses of drugs that have little effect on response rates (Hudzik and McMillan, 1994a).

Depressant drugs have produced mixed results. Pentobarbital and morphine flattened the relative frequency distribution of response durations in one study (Hudzik and McMillan, 1994b) because of a slight increase in the frequency of both shorter and longer response durations. In other studies, these drugs have shown only increases in short response durations (Schulze and Paule, 1991; Ferguson and Paule, 1993). Diazepam was reported to increase the proportion of short response durations (Schulze et al., 1989), whereas in other studies diazepam flattened the distribution of response durations (Hudzik and McMillan, 1994a), an effect similar to pentobarbital and morphine. ⁹-THC increased the relative frequency of shorter response durations under some TRD schedules (Schulze et al., 1988), but flattened the distribution of response durations under other TRD schedules (Schulze et al., 1988; Hudzik and McMillan, 1994b). PCP, which has mixed stimulant-depressant properties, produced only an increase in short response durations that shifted relative frequency distributions leftward (Hudzik and McMillan, 1994b; McMillan et al., 1994).

These conflicting results raise many questions about timing behavior under TRD schedules. Differences in the species used, the schedule parameters, the drug vehicle, routes of administration, the failure to report relative frequency distributions and other procedural aspects make it difficult to systematically compare drug effects on behavior under TRD schedules.

In an attempt to elucidate the role of the schedule parameters in these reported differences, comparisons of behavior of rats under three TRD schedules with different time-duration requirements were conducted. To determine the effects of drugs on the timing behavior of animals under these schedules, MAP, PCP and ⁹-THC were selected as representative abused drugs from different pharmacological classes. Altered time perception has been reported with PCP and ⁹-THC, based on human anecdotal reports (Hollister and Gillespie, 1970; Tinklenberg et al., 1976; Yesavage et al., 1978), and with MAP by use of animal temporal discrimination studies (Meck and Church, 1983; Maricq et al., 1981).

	Methods

Top Abstract Introduction Methods Results Discussion References

Subjects. Twelve male, Sprague-Dawley rats were maintained at 80% (285-315 g) of their free-feeding weights by food presented during the session and postsession feeding. Water was available at all times except during the experimental sessions. Animals were individually housed in suspended stainless steel cages in a colony room (maintained at 70-74°F, illuminated from 6:00 A.M. to 6:00 P.M.) during the initial training period. Later, after a move to new laboratory facilities, all rats were housed singly in standard Plexiglas rat cages under the same environmental conditions. Rats were 9 months of age at the beginning of the experiments, which lasted 7 to 9 months.

Apparatus. Rats were trained and tested in four standard two-lever chambers (model G6312, Gerbrands Corp., Arlington, MA) enclosed in sound-attenuating Gerbrands enclosures (model G7210). A feeder mounted between the levers delivered 97-mg food reinforcer pellets (Noyes Corp., Lancaster, NH) when schedule contingencies had been met. A houselight and a stimulus light consisting of 28-V DC bulbs were mounted in the ceiling of the experimental chamber and above the right lever. A downward force of 15 g closed the contact on the right-hand lever and was defined as a response. Responses on the left-hand lever had no consequences. A continuous tone (Sonalert, model SC628H in series with a 15 K-Ohm resistor) occurred when contacts were closed. Conditioned events in the chambers were controlled and monitored by a Firestar 386 computer with a Med Associates Interface (St. Albans, VT) housed in a separate, but adjoining room.

Drugs and testing. Doses of methamphetamine sulfate (Sigma Chemical Co., St. Louis, MO), phencyclidine hydrochloride and ⁹-tetrahydrocannabinol (National Institutes on Drug Abuse, Bethesda, MD) in this order were given in ascending or descending dose order with random assignment of the rats to a dose order. Both PCP and MAP (0.3, 1.0, 1.7, 3.0 or 5.6 mg/kg) were dissolved in saline and administered intraperitoneally 10 min before sessions in a volume of 1 ml/kg. ⁹-THC, dissolved in ethanol, was evaporated under nitrogen and redissolved in DMSO (Sigma Chemical Co., St. Louis, MO). ⁹-THC (0.3, 1.0, 1.7, 3.0, 4.2 or 5.6 mg/kg) was administered intraperitoneally 1 h before the session in a volume of 0.5 ml/kg. All dose levels were expressed as the salts, except for ^9-THC, which was reported as the free base. Control injections consisted of saline during the MAP and PCP drug series, and DMSO during the ⁹-THC drug series.

Procedure. After all rats learned to lever press, they were randomly assigned to one of the three TRD training schedules and were trained to differentiate response durations by the general method of differential reinforcement as described previously (McMillan and Patton, 1965). Under this training procedure, responses are reinforced based on continuous lever press durations within the required time limits of the schedule. The minimum response duration required for food presentation was progressively incremented (based on a 40% accuracy criterion), then the maximum response duration was progressively decremented until the final minimum and maximum response durations required for food presentation were established. Under the TRD 1-1.3 sec schedule, progressive increments of 0.3 sec with a final 0.1-sec increment of the minimum response duration and decrements of 0.5 sec with a final decrement of 0.2 sec of the maximum response duration were used to achieve the final reinforced response duration of at least 1.0 sec, but less than 1.3 sec. Under the TRD 4-5.2 sec schedule, both progressive 0.5-sec increments and decrements, with a final 0.3-sec decrement, were used to achieve a final reinforced duration of at least 4.0 sec, but less than 5.2 sec. Under the TRD 10-13 sec schedule, both 0.5-sec increments and decrements were used to achieve the final response duration of at least 10 sec but less than 13 sec.

Data analysis. The total number of responses emitted, response rate (responses/sec), accuracy (% reinforced responses) and mean response durations were collected as performance indicators during every session for each rat. The mean values and standard errors for animals under each schedule were calculated for each of these performance measures. ANOVA was used to determine the significance (P < .05) of base-line values compared across schedules. ANOVA was also used to assess the effect of each drug compared with the average of 10 control sessions conducted during the three drug tests for each group of rats. If animals failed to respond at least 25 times within a session, the data were used to calculate only response rates.

(1)

	Results

Top Abstract Introduction Methods Results Discussion References

Base-line performance. Under the TRD 1-1.3 sec, TRD 4-5.2 sec and TRD 10-13 sec schedules, animals reached the initial stability within 50 sessions, 65 sessions and 85 sessions, respectively. Initial stability for each rat was defined as consistent accuracy which varied no more than ±5% for 10 consecutive sessions, after completion of training procedures. At the beginning of the experiments, all rats under the TRD 1-1.3 sec schedule had accuracy levels equal to or greater than 45%, whereas the accuracy of all rats under the 4-5.2 sec and 10-13 sec schedules was 55% or greater. Base-line levels of accuracy continued to increase slightly under some schedules after the initial stability criterion was reached. However, an asymptote was reached for all groups at which continued daily training had no additional effects. Before each drug series, consistent accuracy levels (±5%) during the previous five base-line trials were required before drugs were administered and drug effects assessed.

View this table: [in this window] [in a new window]   TABLE 1 Effect of MAP, PCP and 9-THC on performance indicators for TRD schedules with differing timing requirements

View larger version (27K): [in this window] [in a new window]   Fig. 1.   Comparison of base-line cumulative frequencies of distributions and relative frequency distributions of response durations for TRD schedules with differing timing requirements. Abscissa: Response durations in 0.1-, 0.4- or 1.0-sec time bins. Ordinate: Percentage of total response durations in each bin (left) or cumulative number of response durations in each bin as a percentage of the total number of response durations (right). Shaded bins show reinforced response durations. The frequency of each bin represents a mean based on 10 observations in each of four rats. Brackets show ±1 standard error.

General performance measures after drug administration. Table 1 shows the effects of MAP, PCP and ⁹-THC on the various measures of performance under all TRD schedules. MAP caused a dose-dependent decrease in accuracy under all TRD schedules, with greater decreases in accuracy occurring with increasing time duration requirements. For MAP, statistically significant effects occurred under the TRD 1-1.3 sec schedule (1.7 and 3 mg/kg), under the TRD 4-5.2 sec schedule (1.7 and 3.0 mg/kg) and under the TRD 10-13 sec schedule (all doses tested). There was no effect on mean response duration under the TRD 1-1.3 sec schedule, whereas large decreases in mean response durations occurred under either the TRD 4-5.2 sec schedule (5-fold) and the TRD 10-13 sec schedule (10-fold). Response rates were generally not affected by MAP until dose levels were reached that nearly eliminated responding.

MAP effects on distribution of response durations under the TRD 1-1.3 sec schedule. Figure 2 shows cumulative frequency curves which were generated from the relative frequency data with inserts showing the relative frequency distributions. At the 0.3 mg/kg dose, there was an increase in the relative frequency of response durations that were too short to produce the reinforcer and the peak of the distribution shifted one bin to the left (fig. 2A). As the dose of MAP increased, the relative frequency of response durations that were too short, as well as those that were too long to produce the reinforcer increased, which resulted in a progressive flattening of the response distribution (fig. 2D).

View larger version (42K): [in this window] [in a new window]   Fig. 2.   Effects of MAP on the cumulative frequency of response-duration distributions under the TRD 1-1.3 sec schedule. Abscissa: Response durations in 0.1-sec time bins. Ordinate: Cumulative number of responses in each bin as a percentage of the total number of responses. Each point for MAP curves represents a mean of single observations in each of four rats. Each point for the control curve represents a mean based on 10 observations in each of four rats. Brackets show ±1 standard error. * Signifies a significant difference from the control curve. Doses are as indicated in the key. The inserts show the relative frequency distributions for the cumulative curves as indicated in the key. Insert Ordinate: Percentage of total responses in each bin. Shaded bins show reinforced response durations.

MAP effects on distribution of response durations under the TRD 4-5.2 sec schedule. Figure 3 shows the cumulative frequency curves of response durations based on the relative frequencies with increasing doses of MAP under the TRD 4-5.2 sec schedule. With each increasing dose, the cumulative frequency curves were shifted upward and to the left without any change in asymptote; this reflected shifts of the entire distribution toward short response durations as the dose of MAP increased. At the 1.0 mg/kg dose, an increase in the frequency of responses of short duration (<1.2 sec in duration) began to occur although there was little effect on the mode of the relative frequency distribution (fig. 3B). This is reflected in the cumulative frequency curves with the large increase in short duration responses (bins 1 and 2) causing an initial increase in the slope of the curve before a further vertical rise. However, with a quarter log increase in dose to 1.7 mg/kg, a large increase in short responses occurred. At the highest dose, more than 50% of the responses are short (<1.6 sec in length) causing the curve to shift upward and to the left. These effects of MAP under the TRD 4-5.2 sec schedule were much larger than those under the TRD 1-1.3 sec schedule. The linear portions of the MAP curves were significantly different from the control curve at all dose levels of MAP.

View larger version (48K): [in this window] [in a new window]   Fig. 3.   Effects of MAP on cumulative frequency distributions of response durations under the TRD 4-5.2 sec schedule. Abscissa: Response duration time bins in 0.4 sec. Other details as in figure 2.

MAP effects on response-duration distribution under the TRD 10-13 sec schedule. Figure 4 shows the cumulative frequency curves based on the relative response-duration distributions under the TRD 10-13 sec schedule. The cumulative frequency distributions again showed a shift upward and to the left, and at lower doses this shift occurred with little change in the asymptotic portion of the curve. At the 0.3 mg/kg dose, there was an increase in the relative frequency of short response durations in nonreinforced bins, shifting the cumulative curve leftward. This effect was more pronounced with higher doses (fig. 4, B and D) with a high relative frequency of short responses causing cumulative curves to shift upward and leftward. At the 1.7 mg/kg dose, more than 50% of the responses had occurred by bin 2 (<2 sec in duration), and at the 3.0 mg/kg dose (fig. 4D), 67% of all responses were less than 1 sec in duration. At that dose, the asymptote was reached in bin 4, which indicated that more than 90% of the responses were less that 4.0 sec in duration and almost no responses were long enough to fall in the reinforcer window. All doses of MAP produced cumulative curves in which the linear portions were significantly different from the linear portion of the control curve.

View larger version (48K): [in this window] [in a new window]   Fig. 4.   Effects of MAP on cumulative frequency distributions of response durations under the TRD 10-13 sec schedule. Abscissa: Response duration time bins in 1.0 sec. Other details as in figure 2.

PCP effects on distribution of response durations under the TRD 1-1.3 sec schedule. Figure 5 shows the cumulative response frequency curves for PCP. The distributions shifted to the left with shorter response durations occurring at the 0.3 mg/kg dose level (fig. 5A). This trend increased with increasing doses of PCP. At the 3.0 mg/kg dose, there was an increase in relative frequency across all the nonreinforced bins (fig. 5D) to the left of the reinforcer window (bins 1-10), which caused a decrease in the cumulative curve slope. There was a rather large shift to the left between the 1.7 and the 3.0 mg/kg dose levels corresponding with the rather steep dose-effect curve that exists with PCP. There were no corresponding long-duration responses as with the TRD 1-1.3 sec schedule for MAP, and therefore little change in the asymptote with any dose. A significant difference between the linear portion of the control curve and the PCP curves existed at all doses.

View larger version (42K): [in this window] [in a new window]   Fig. 5.   Effects of PCP on cumulative frequencies of distributions of response durations under the TRD 1-1.3 sec schedule. Abscissa: Response duration time bins in 0.1 sec. Each point for PCP curves represents a mean of single observations in each of four rats. Other details as in figure 2.

PCP effects on response-duration distribution under the TRD 4-5.2 sec schedule. The cumulative frequency curves (fig. 6) showed an increasing shift to the left with increasing doses of PCP. A feature of this drug's effect on the relative frequency of the response-duration distribution was an increase in short-duration responses with a bimodal pattern starting at the 1.0 mg/kg level (fig. 6B). At doses of 1.0 and 1.7 mg/kg, a bimodal pattern produced by an increase in relative frequencies in early bins caused the early portion of the cumulative curve to shift leftward (curves B and C). At the highest dose, the curve had shifted far to the left and the frequency of short-response durations was more similar across all the first 10 bins (curve D). The final effect was that of a fairly uniform frequency of short response durations with few long enough to be reinforced. At doses of 1.0, 1.7 and 3.0 mg/kg the linear portions of the PCP curves were significantly different from the linear portion of the control curve. Again, the effects were more pronounced under the TRD 4-5.2 sec than under the TRD 1-1.3 sec schedule.

View larger version (47K): [in this window] [in a new window]   Fig. 6.   Effects of PCP on cumulative frequency distributions of response durations under the TRD 4-5.2 sec schedule. Abscissa: Response duration time bins in 0.4 sec. Each point for PCP curves represents a mean of single observations in each of four rats. Other details as in figure 2.

PCP effects on response-duration distributions under the TRD 10-13 sec schedule. The cumulative frequency curves (fig. 7) again showed a shift to the left at low doses of 1.0 and 1.7 mg/kg PCP. Also, as under the TRD 4-5.2 sec schedule at the highest dose, few response durations were long enough to be reinforced (fig. 7D), and the distribution in the first 10 bins represented a fairly uniform overestimation of time (responses too short). A statistically significant difference from the control curve existed in the linear portions of the PCP curves at doses of 1.0, 1.7 and 3.0 mg/kg.

View larger version (45K): [in this window] [in a new window]   Fig. 7.   Effects of PCP on cumulative frequency distributions of response durations under the TRD 10-13 sec schedule. Abscissa: Response duration time bins in 1.0 sec. Each point for PCP curves represents a mean of single observations in each of three rats (nos. 422, 424 and 428). Each point for the control curve represents a mean based on 10 observations in each of the same three rats. Other details as in figure 2.

⁹-THC effects on performance under the TRD 1-1.3 sec schedule. The cumulative frequency curves (fig. 8) emphasized the lack of effect on the response-duration distributions. Although the 1.7 mg/kg dose produced a slight shift to the right, which indicated a slight increase in longer responses, none of the curves were significantly different from control in the linear regions of the curves.

View larger version (32K): [in this window] [in a new window]   Fig. 8.   Effects of 9-THC on cumulative frequencies of distributions of response durations under the TRD 1-1.3 sec schedule. Abscissa: Response duration time bins in 0.1 sec. Each point for 9-THC curves represents a mean of single observations in each of four rats. Other details as in figure 2.

⁹-THC effects on performance under the TRD 4-5.2 sec schedule. The cumulative frequency curves (fig. 9) showed little effect of the various doses, similar to the lack of effect under the TRD 1-1.3 sec schedule. At the 0.3 mg/kg dose, there was a slight shift to the left in the upper part of the curve, which indicated a slight increase in shorter responses and the loss of some longer responses in or near the reinforcer window. The 1.7 mg/kg dose caused the slope to be altered in the early portion of the curve characteristic of an increase in short responses in the early bins and a slight decrease in asymptote caused by the long responses. Both the 0.3 and 1.7 mg/kg curves were significantly different from the control curve when the linear portions of the ⁹-THC curves were analyzed. None of the other curves were significantly different from the control curve.

View larger version (34K): [in this window] [in a new window]   Fig. 9.   Effects of 9-THC on cumulative frequency distributions of response durations under the TRD 4-5.2 sec schedule. Abscissa: Response duration time bins in 0.4 sec. Each point for 9-THC curves represents a mean of single observations in each of four rats. Other details as in figure 2.

⁹-THC effects on performance under the TRD 10-13 sec schedule. The cumulative frequency curves (fig. 10) showed a shift to the left, which indicated shorter responses at even the lowest dose. The 4.2 mg/kg dose shifted the curve greatly to the left with most of the responses occurring in only a few early bins indicated by the vertical slope and the earlier point of asymptote. More than 90% of the responses are less than 6 sec in duration. The linear portions of all the ⁹-THC curves were significantly different from control curve at all doses.

View larger version (40K): [in this window] [in a new window]   Fig. 10.   Effects of 9-THC on cumulative frequency distributions of response durations under the TRD 10-13 sec schedule. Abscissa: Response duration time bins in 1.0 sec. Each point for 9-THC curves represents a mean of single observations in each of three rats (nos. 422, 424 and 427). Each point for control curve represents a mean based on 10 observations in each of the same three rats. Other details as in figure 2.

	Discussion

Top Abstract Introduction Methods Results Discussion References

All of the drugs in the present study have been reported anecdotally to alter time perception and to produce overestimation of the passage of time in controlled studies (Hicks et al., 1984; Karniol et al., 1975; Yesavage and Freeman, 1978; Meck, 1983). The present experiments showed that the effects of these drugs on responding under TRD schedules differed from each other and that the differences depended not only on the drug and the dose level, but also on the time requirements of the TRD schedule. Therefore, a simple characterization of these drugs as drugs that alter time perception is overly simplistic.

TRD schedules with longer timing requirements produced behaviors that were more sensitive to drug effects than did the TRD schedules with shorter timing requirements. For example, the TRD 10-13 sec schedule was the only schedule in which effects of ⁹-THC were apparent. Furthermore, the effects of both PCP and MAP appear to be larger under the two longer TRD schedules than under the TRD 1-1.3 sec schedule (compare figs. 3 and 4 with fig. 2 and figs. 6 and 7 with fig. 5). The reasons for the greater sensitivity to these drugs under the TRD schedules with longer timing requirements is not clear. One possibility is the control performance. As the minimum lever press required for food presentation increased, the accuracy increased under control conditions without an increase in the variability. Perhaps these base-line differences in accuracy influenced the size of the drug effect.

The effects of each drug also depended on the schedule requirements. The most striking example was ⁹-THC, in which few effects were observed under the TRD 1-1.3 and TRD 4-5.2 sec schedules, but large effects were observed under the TRD 10-13 sec schedule. Schulze et al. (1988) studied the effects of ⁹-THC in rhesus monkeys responding under a TRD 10-14 sec schedule and reported effects similar to those observed in this study. It is unlikely that tolerance, which may develop extremely rapidly to ⁹-THC with a single dose, influenced the effects under some TRD schedules in these studies and not others. Each schedule was performed by a different group of animals and animals within each group were randomly assigned an ascending or descending dose order. The observed greater effects of ⁹-THC under schedules with longer time durations and little effect under the other schedules with shorter time duration requirements were consistent among the rats within each group.

MAP decreased accuracy as a function of dose under all TRD schedules, but the effects were different under TRD schedules with different timing requirements for food presentation. Under the TRD 1-1.3 sec schedule, MAP flattened the relative frequency distribution, but under the TRD 4-5.3 sec and TRD 10-13 schedules, the relative frequency distribution shifted to the left with high doses producing only responses of short duration. The flattening of the relative frequency distributions by MAP under the TRD 1-1.3 sec schedule is consistent with reports by Hudzik and McMillan (1994b) and McMillan et al. (1994). These investigators reported a flattening of the relative response-duration distributions after MAP, especially at the 3.0 mg/kg dose. The shift of the response-duration distributions to the left after amphetamines that was seen under the TRD 10-14 sec schedule has also been observed when rhesus monkeys served as subjects (Schulze and Paule, 1990).

That a drug such as MAP might produce differential effects under the TRD 1-1.3 sec schedule than under TRD schedules with longer timing requirements is not too surprising. The TRD 1-1.3 sec schedule requires that a precision motor response be made. Release of the lever within a 300-msec period may require motor responses not required by the longer TRD schedules. The coordinated motor movements required by the TRD 1-1.3 sec schedule may be very different from the requirements of longer TRD schedules in which motor precision is less likely to be important. Failure to release the lever promptly might be more likely to cause response durations longer than those reinforced under the 300-msec reinforced time span. This coupled with the short-duration responses seen under all three TRD schedules could produce a flattening of the relative frequency distribution. Such effects would seem less likely to occur under the wider reinforcer window associated with TRD schedules that required longer lever holds for food presentation. This suggests that although TRD schedules with longer timing parameters require behaviors that are more sensitive to the effects of the drugs, the shorter timing parameters associated with the TRD 1-1.3 sec schedule may involve quite different behaviors that are differentially affected by the drugs used in this study.

In contrast to MAP and ⁹-THC, PCP produced similar effects across all TRD schedules, although the effects were smaller at TRD 1-1.3 sec than for the longer TRD schedules. The shift in the relative frequency distribution to the left with a fairly equal distribution over the first 10 bins was similar to that reported in previous studies with the TRD 1-1.3 sec schedule (Hudzik and McMillan, 1994b) in which a similar effect was observed. It is not clear why PCP effects appear to depend less on the time requirements of the TRD schedules than do those of MAP and ⁹-THC.

Although relatively few investigators have studied the effects of drugs on responding under TRD schedules, there are several studies of the effects of these drugs on responding under DRL schedules at similar time parameters. Both TRD and DRL schedules may be considered to be time production schedules, in that the TRD schedule requires an animal to make a response (lever press) of a specified duration to produce the reinforcer, whereas the DRL schedule requires the animal to terminate a time period of a specified duration with a response.

Amphetamines generally shift IRT distributions toward shorter IRTs under DRL 10-sec schedules (Wenger and Wright, 1990; Lucki and DeLong, 1983), although a few studies have found increases in long IRTs after amphetamines (Balster and Baird, 1979). PCP has been reported to produce variable results on responding under DRL 10-sec schedules. In most studies PCP has shifted the IRT distributions toward shorter times (Sanger, 1992), similar to the effects observed in the present studies, although there are also reports of flattening of the IRT distributions (Freeman et al., 1984; Balster and Baird, 1979). ⁹-THC also usually produces shifts to the left in the IRT distributions under DRL 10-sec schedules (Ferraro et al., 1971; Galbicka et al., 1980; Manning, 1973), similar to the findings in the present experiments. Thus in general, TRD 10-13 sec schedules and DRL 10-sec schedules do not differ much in the effects that these drugs produce on behavior maintained under these schedules. All of the drugs are most frequently characterized by shifts toward shorter IRTs or shorter response durations under these schedules.

DRL schedules requiring minimum IRTs shorter than 10 sec have rarely been used in behavioral pharmacology. An exception is the work of McClure and McMillan (1997) who studied the effects of these same drugs under DRL schedules with the same timing requirements as those for the present TRD schedules in a direct attempt at comparison. The relative frequency distributions with PCP under TRD 4-5.2 and TRD 10-13 sec schedules were similar to those produced under similar DRL schedules. However, under the TRD 1-1.3 sec schedule, PCP shifted responding toward short-response durations, whereas under the DRL 1-1.3 sec schedule, the relative frequency distribution shifted toward long interresponse times. Sanger (1992) and Sanger and Jackson (1989) have also reported a shift in the relative frequency distributions to the left with a flat relative frequency distribution of short responses in studies with rats under a DRL 15-sec schedule. However, Balster and Baird (1979) and Freeman et al. (1984) found more extreme effects in mice responding under a DRL 10-sec schedule with most response durations occurring in the first few bins, although these effects were obtained only at a high dose (19.0 mg/kg).

With MAP, the extreme leftward shift of the relative frequency distribution under TRD 4-5.2 sec and TRD 10-13 sec schedules differed from the flatter distribution of early responses present under the DRL 4-5.2 sec and 10-13 sec schedules (McClure and McMillan, 1997). This effect under the DRL 10-13 sec schedule (McClure and McMillan, 1997) is not unlike the pattern that is produced with MAP or d-amphetamine under many DRL schedules with time parameters of 10 sec or slightly longer (Balster and Baird, 1979; Wenger and Wright, 1990). Under schedules requiring longer IRTs (Sidman, 1955; Pradhan and Dutta, 1970; Sanger et al., 1974; Hodos et al., 1962; Adam-Carriere et al., 1978; Seiden et al., 1979; Levine et al., 1980) MAP produces a leftward shift of the relative frequency distribution that is more extreme, like the effect of MAP under the TRD 4-5.2 sec or TRD 10-13 sec schedules.

⁹-THC also produced a greater effect on responding under TRD schedules than under comparable DRL schedules. Although little or no effect on responding occurred at 1-1.3 sec or 4-5.2 sec time parameters under either TRD or DRL schedules, relative frequency distributions showed leftward shifts under both 10-13 sec schedules. At these longer timing requirements, ⁹-THC produced a more prominent shift to the left under the TRD 10-13 sec schedule than under the DRL 10-13 sec schedule (McClure and McMillan, 1997). The effect of ⁹-THC under the DRL 10-13 sec schedule was remarkably similar to the effects of ⁹-THC on responding under all DRL patterns reviewed for comparison (Ferraro et al., 1971, 1972; Manning, 1976; Galbicka et al., 1980). In all DRL studies surveyed, if the lower limit of the reinforced window is 10 sec or longer, all distributions retain the bell-shape distribution but shifted to the left toward shorter IRTs. The more extreme leftward shift under the TRD 10-13 sec schedule indicated a greater effect of ⁹-THC under this TRD schedule than that under the other DRL schedules reviewed. Therefore, the contrasting effects of MAP and ⁹-THC under TRD and DRL schedules suggest that the two types of timing schedules may require different behavioral processes.

The comparison between TRD and DRL schedules and the effects of drugs on responding maintained by these schedules suggests that differential drug effects can occur for responding under TRD and DRL schedules even when the time duration requirements for the schedules are identical. Because both TRD and DRL schedules are time production tasks, they must differ from each other is some other way if drugs are to have differential effects on behavior under these schedules. The base-line accuracy rates for the TRD schedules were much higher and the standard errors were much lower than those under the comparable DRL schedules (McClure and McMillan, 1997). These base-line differences may contribute to the differential effects of some drugs. Because TRD schedules require the animal to maintain contact with the lever, they restrict the animal's movement relative to DRL schedules where the animal is free to move about the cage during the IRT. This would be a logical place to begin to look for the determinants of the differential sensitivity to the drug effects.