G Model
ARTICLE IN PRESS
BEPROC 3067 1–5
Behavioural Processes xxx (2015) xxx–xxx
Contents lists available at ScienceDirect
Behavioural Processes journal homepage: www.elsevier.com/locate/behavproc
1
Short report
3
Differential effects of massed and spaced training on place and response learning: A memory systems perspective
4
Jeffrey C. Wingard b , Jarid Goodman a , Kah-Chung Leong a , Mark G. Packard a,b,∗
2
5 6
a b
Department of Psychology and Institute for Neuroscience, Texas A&M University, USA Department of Psychology, Yale University, USA
7
8 15
a r t i c l e
i n f o
a b s t r a c t
9
14
Article history: Received 31 March 2015 Received in revised form 8 May 2015 Accepted 1 June 2015 Available online xxx
16
1. Introduction
10 11 12 13
Studies employing brain lesion or intracerebral drug infusions in rats have demonstrated a double dissociation between the roles of the hippocampus and dorsolateral striatum in place and response learning. The hippocampus mediates a rapid cognitive learning process underlying place learning, whereas the dorsolateral striatum mediates a relatively slower learning process in which stimulus-response habits underlying response learning are acquired in an incremental fashion. One potential implication of these findings is that hippocampus-dependent learning may benefit from a relative massing of training trials, whereas dorsal striatum-dependent learning may benefit from a relative distribution of training trials. In order to examine this hypothesis, the present study compared the effects of massed (30 s inter-trial interval; ITI) or spaced (30 min ITI) training on acquisition of a hippocampus-dependent place learning task, and a dorsolateral striatum-dependent response task in a plus-maze. In the place task rats swam from varying start points (N or S) to a hidden escape platform located in a consistent spatial location (W). In the response task rats swam from varying start points (N or S) to a hidden escape platform located in the maze arm consistent with a body-turn response (left). In the place task, rats trained with the massed trial schedule acquired the task quicker than rats trained with the spaced trial schedule. In the response task, rats trained with the spaced trial schedule acquired the task quicker than rats trained with the massed trial schedule. The double dissociation observed suggests that the reinforcement parameters most conducive to effective learning in hippocampus-dependent and dorsolateral striatum-dependent learning may have differential temporal characteristics. © 2015 Published by Elsevier B.V.
Beneficial effects of spaced training over massed training schedules have been observed in various learning tasks across a range 18 of species. In lower animals this phenomenon has been reliably 19 20 demonstrated in various Pavlovian conditioning paradigms, includ21 ing for example eyelid conditioning (e.g., Salafia et al., 1973), taste 22 aversion conditioning (Domjan, 1980), and conditioned leg move23 ments (O’Brien and Packham, 1973). However, rare exceptions to 24 the facilitative effects of spaced training on learning do exist. In 25 rats, for instance, massed training results in superior acquisition of 26 “place” learning in both an appetitive plus-maze task (Thompson 27 and Thompson, 1949) and visuospatial discrimination task (Honey, 28Q1 1996). 17
∗ Corresponding author at: Department of Psychology, Texas A&M University, College Station, TX 77843, USA. Tel.: +1 979 845 9504; fax: +1 979 845 4727. E-mail address: [email protected] (M.G. Packard).
Such exceptions may potentially be explained within the context of multiple memory systems. According to this view, mammalian memory is organized in relatively independent brain systems that mediate different types of memory (for reviews see Packard, 2001; White and McDonald, 2002; Packard and Goodman, 2012; White et al., 2013). Several experiments investigating the neuroanatomical bases of multiple memory systems have revealed dissociable mnemonic roles for the hippocampus and dorsolateral striatum (DLS), in “cognitive” and stimulus-response “habit” memory, respectively (e.g., Packard et al., 1989; Packard and McGaugh, 1992; McDonald and White, 1993; Kesner et al., 1993). In addition to differing in their information content, the types of memory mediated by the hippocampus and DLS may differ in their temporal characteristics. The hippocampus is hypothesized to mediate a rapid form of cognitive learning, whereas the DLS mediates a habitual form of learning in which stimulus-response habits accrue in an incremental or gradual fashion. Thus, in “dual-solution” tasks in which both systems can provide an adequate learned solution, the relative contribution of the hippocampus and DLS changes over
http://dx.doi.org/10.1016/j.beproc.2015.06.004 0376-6357/© 2015 Published by Elsevier B.V.
Please cite this article in press as: Wingard, J.C., et al., Differential effects of massed and spaced training on place and response learning: A memory systems perspective. Behav. Process. (2015), http://dx.doi.org/10.1016/j.beproc.2015.06.004
29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47
G Model BEPROC 3067 1–5
J.C. Wingard et al. / Behavioural Processes xxx (2015) xxx–xxx
49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86
the course of training. Specifically, hippocampus-dependent cognitive learning dominates early in training, whereas DLS-dependent habit learning dominates after extended training (e.g., Packard and McGaugh, 1996). One implication of the idea that the hippocampus and DLS mediate rapid and slow forms of learning, respectively, is that the former memory system may benefit from a relative massing of training trials, whereas the latter system may benefit from a relative distribution of training trials. This is consistent with the observation that the DLS-dependent memory system typically comes “online” later in training than the hippocampus-dependent memory system (Ritchie et al., 1950; Hicks, 1964; Packard and McGaugh, 1996; Chang and Gold, 2003), potentially making spaced training more conducive to response learning. The relative effectiveness of massed and spaced training protocols in place and response learning may also be interpreted within the context of classical learning theories (Ebbinghaus, 1913; McGeoch, 1932; Hull, 1943; Roberts and Grant, 1976), as well as the observation that hippocampus- and DLS-dependent memory systems sometimes compete for control of learning (Poldrack and Packard, 2003). The present study examined whether massed or spaced training differentially benefit the forms of learning mediated by the hippocampus and DLS. Accordingly, rats received either massed training (i.e., short inter-trial interval) or distributed (i.e., long inter-trial interval) training in two water plus-maze tasks, one task that required hippocampus-dependent “place” learning and another task that required DLS-dependent “response” learning. Importantly, the place and response learning tasks included the same motivational, motoric, and sensory requirements and were also conducted in the same maze environment. These tasks only differed in terms of the learning requirement, and thus any difference between massed and spaced training protocols in the place and response tasks may be more likely attributed to the protocols differentially affecting place and response learning, as opposed to an effect on non-mnemonic processes. It was hypothesized that massed training would result in superior learning in the hippocampus-dependent place task, whereas spaced training would result in superior learning in the DLS-dependent response task.
87
2. General methods
88
2.1. Subjects
95
Subjects were 35 experimentally naïve male Charles River Long–Evans rats (weighing 300–350 g at arrival), individually housed in a climate-controlled vivarium with food and water provided ad libitum. After arrival, animals were given a one-week acclimation period before experiments began. Animals were on a 12:12-h light:dark cycle (lights on at 7 a.m.), and experiments were conducted during the light phase of this cycle.
96
2.2. Apparatus
89 90 91 92 93 94
97 98 99 100 101 102 103 104 105 106 107
Rats were trained in a black circular water maze (1.83 m diameter, 0.58 m in height) into which a clear Plexiglas plusmaze (59.69 cm height, arm-width of 25.4 cm, and arm-length of 43.18 cm) was inserted. The maze was filled to a water level of 20 cm. The water temperature was held constant at 25 ◦ C throughout training. Extra-maze visual cues of various geometric shapes were placed on square dark green curtains surrounding the water maze (none of the extra-maze cues were placed in a spatially congruent/proximal position with the ends of the maze arms). An invisible black Plexiglas escape platform (11 × 14 × 19 cm) was located in one arm of the maze and submerged 1 cm below water
Place Learning 100 90
Percent Correct
2 48
ARTICLE IN PRESS
80 70 60 50
Spaced ITI
40
Massed ITI
30 20 0
1
2
3
4
5
Day Fig. 1. Effect of training with a spaced or massed inter-trial interval (ITI) on the acquisition of a single-solution place water plus-maze task. Relative to animals given spaced training (30 min ITIs), animals given massed training (30 s ITIs) displayed a greater percentage of correct arm entries over the course of training in the place learning task.
level. Swimming behavior was observed and recorded by an experimenter standing outside the curtains and using an overhead video camera system and monitor. 2.3. Behavioral procedures Behavioral training was conducted over 5 consecutive days, and each day consisted of 6 trials (i.e. swims) in either a place or response plus-maze task. An inter-trial interval (ITI) of 30 min was used for rats given spaced training, and an ITI of 30 s was used for rats given massed training. The rats started from either the north or south arms. On odd days, the sequence of start positions was NSSNNS; and on even days, SNNSSN. On a given trial, a clear Plexiglas barrier blocked access to the arm directly opposite the start arm. For the place task, rats (spaced training, n = 7; massed training, n = 7) were trained to swim to a submerged escape platform that was always located in the same goal arm of the maze (west). In the response task (spaced training, n = 11; massed training, n = 10), the escape platform was always located in the maze goal arm to the left of the start arm, requiring the rat to learn a specific body-turn response to approach the escape platform (i.e. when the start arm was north, the platform was in the west arm; and when the start arm was south, the platform was in the east arm). After reaching the escape platform, rats remained on it for 10 s. Rats that failed to find the hidden platform within 60 s were manually guided to it. The experimenter recorded the number of errors made on each trial. An error was defined as swimming a full body length into the non-goal maze arm. After removal from the escape platform, rats were returned to their cages located on shelving adjacent to and outside of the behavioral testing room for the assigned ITI. After a rat finished a training session, the rat was again returned to its holding cage until all rats were finished training for the day. In this way, animals in the spaced and massed conditions spent an equal amount of time in the holding cage on the table adjacent to the maze room for each day of training. 3. Results Fig. 1 depicts the mean percentage of correct responses during training in the place task. A two-way one-repeated measure ANOVA (group × training day) computed on percent correct data for all five days revealed a significant main effect of group [F (1, 12) = 7.1, p < 0.05]. Main effect of training day [F (4, 48) = 2.4, n.s.],
Please cite this article in press as: Wingard, J.C., et al., Differential effects of massed and spaced training on place and response learning: A memory systems perspective. Behav. Process. (2015), http://dx.doi.org/10.1016/j.beproc.2015.06.004
108 109 110
111
112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140
141
142 143 144 145 146
G Model
ARTICLE IN PRESS
BEPROC 3067 1–5
J.C. Wingard et al. / Behavioural Processes xxx (2015) xxx–xxx
Response Learning 100
Percent Correct
90 80 70 60 50
Spaced ITI
40
Massed ITI
30 20 0
1
2
3
4
5
Day Fig. 2. Effect of training with a spaced or massed inter-trial interval (ITI) on acquisition of a single-solution response water plus-maze task. Relative to animals given massed training (30 s ITIs), animals given spaced training (30 min ITIs) displayed a greater percentage of correct arm entries over the course of training in the response learning task.
159
and the training day × group interaction [F (4, 48) = 0.12, n.s.] were not significant. These data indicate that in the place learning task rats receiving massed training displayed superior learning relative to rats receiving spaced training. Fig. 2 depicts the mean percentage of correct responses during training in the response task. A two-way one-repeated measure ANOVA (group × training day) computed on percent correct data for all five days revealed a main effect of group [F (1, 9) = 4.6, p < 0.05], and training day [F (4, 76) = 14.9, p < 0.001]. The group × training day interaction was not significant [F (4, 76) = 0.67, n.s.]. These data indicate that in the response learning task rats receiving spaced training displayed superior learning relative to the rats receiving massed training.
160
4. Discussion
147 148 149 150 151 152 153 154 155 156 157 158
161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187
The findings indicate a double dissociation of the effects of massed and spaced training on place and response learning. Specifically, that massed training, relative to spaced training, enhances acquisition in a “place” plus-maze task, whereas spaced training enhances acquisition in the “response” task. These findings are consistent with a similar study that used appetitive versions of the place and response plus-maze tasks (Thompson and Thompson, 1949). In view of extensive evidence from neurobehavioral studies indicating that place and response learning in the plus-maze critically depend on the hippocampus and DLS, respectively (Packard and McGaugh, 1996; Packard, 1999; Chang and Gold, 2003; Yin and Knowlton, 2004), massed training may in general benefit learning mediated by the hippocampus, whereas spaced training may benefit learning mediated by the DLS. Given the aversive nature of the water plus-maze tasks and the role that stress/anxiety plays in multiple memory systems (Packard, 2009), it should be noted that stress/anxiety might have contributed to the effects of massed and spaced training protocols in the present study. However previous findings indicating that similar effects may be observed in appetitive versions of the place and response plus-maze tasks (Thompson and Thompson, 1949) suggest that the effects of massed and spaced training protocols may persist regardless of the stressful nature of the learning task. The enhancing effect of spaced training on response learning is consistent with a large body of evidence showing distributed practice improves learning in motor/procedural-like tasks in lower animals (e.g. Salafia et al., 1973; Domjan, 1980) and humans (e.g.,
3
Abrams and Grice, 1976; McBride and Payne, 1979; Glenberg, 1992). On the other hand, the finding that a relative massing of trials benefits spatial learning appears at first glance inconsistent with previous research, which has shown in contrast a benefit of spaced training in various spatial tasks (e.g., Goodrick, 1973; Cohen et al., 1994), including spatial learning in the water maze (Morris and Doyle, 1985; Kraemer and Randall, 1995; Spreng et al., 2002; Commins et al., 2003). One possible reason for this inconsistency is that previous studies used different parameters to define massed and spaced training conditions. Massed training in the water maze, for instance, has typically been conducted over 1 day, whereas spaced training has consisted of the same number of trials, but with the trials distributed over multiple days (e.g., Morris and Doyle, 2003). Thus, spaced training in such previous studies may have permitted long-term consolidation processes (like those that purportedly occur during sleep; Wilson and McNaughton, 1994; Albuoy et al., 2013), which otherwise may have been prevented by a 1-day massed training protocol. The design of the present study avoided this potential confound by ensuring that the number of training days was equal between massed and spaced conditions, whereas the ITI within a daily training session differed between the groups. In this way, the relative impairing effect of spaced training observed in the present spatial task is consistent with other previous work showing that longer ITIs impair acquisition and retention in a variety of spatial tasks (Thompson and Thompson, 1949; Olton et al., 1979; Bureˇsová, 1980; Panakhova et al., 1984; Bolhius et al., 1985). With regard to the mechanism(s) that may underlie the present findings, several principles in classical learning theory may be proposed, including the trace decay hypothesis (Ebbinghaus, 1913; Roberts and Grant, 1976), interference theory (McGeoch, 1932), and reactive inhibition (Hull, 1943). The trace decay hypothesis suggests that after an event is terminated, the memory trace for that event begins to deteriorate over time. In the Morris water maze, increasing the intertrial interval between an initial training trial and subsequent test trial is associated with greater escape latencies (Panakhova et al., 1984). In the present study, rats receiving distributed training in the spatial task may have experienced greater memory decay over the 30-min intertrial intervals, relative to the 30-s intervals, and for this reason displayed slower acquisition. However, trace decay fails to account for the beneficial effect of distributed training in the response task. In this case, findings in the response task may be partly explained by interference theory. Interference putatively occurs when one memory trace interferes with the retrieval of another trace (McGeoch, 1932), and this interference becomes more likely when the two traces are acquired in the same environment (Bilodeau and Schlosberg, 1951; Underwood, 1957). During acquisition of the response task, rats may in the beginning encode both the spatial location of the platform in a given trial and also the turning response used to reach the platform (consistent with the hypothesis of “multiple parallel memory systems”; White and McDonald, 2002). However, over the course of training in the response task, memory of the platform’s spatial location in a previous trial may interfere with the retrieval of a turning response that sometimes leads to a different spatial location. This would result in occasional errors. Nevertheless, considering that interference from a previous memory may dissipate over time (Underwood, 1957), longer ITIs could potentially prevent interference from the old spatial memory, permitting uninhibited retrieval of the turning response and, thus, fewer errors over the course of training in the response task. Another mechanism that may underlie the benefit of distributed training in the “response” plus-maze task is reactive inhibition (Hull, 1943). Reactive inhibition refers to a hypothesized “negative drive” that an organism experiences after performing a rewarded response; this negative drive thus makes the animal less likely
Please cite this article in press as: Wingard, J.C., et al., Differential effects of massed and spaced training on place and response learning: A memory systems perspective. Behav. Process. (2015), http://dx.doi.org/10.1016/j.beproc.2015.06.004
188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253
G Model BEPROC 3067 1–5 4 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318
ARTICLE IN PRESS J.C. Wingard et al. / Behavioural Processes xxx (2015) xxx–xxx
to repeat that same response for a brief time. A purported consequence of reactive inhibition is the tendency for an animal to spontaneously alternate choice behavior in the T-maze (Solomon, 1948), and in this way reactive inhibition may lead to incorrect turns in learning situations that require the same turning response across trials (Thompson and Thompson, 1949), as in the present “response” plus-maze task. However, as reactive inhibition may gradually diminish over time (Hull, 1943), longer intertrial intervals could reduce this tendency to alternate (Heathers, 1940; Zeaman and House, 1951) and therefore, in the present study, could have prevented incorrect turns in the response task. An additional consideration is that the memory systems mediating place and response learning in these tasks can sometimes “compete” for control of learning (Poldrack and Packard, 2003). An implication of this competitive interaction is that by impairing one memory system, learning that is mediated by the other “intact” system may be facilitated. For instance, lesions of the hippocampal system facilitate acquisition in a variety of dorsal striatumdependent response tasks (Packard et al., 1989; McDonald and White, 1993; Schroeder et al., 2002). A similar process may emerge with the mere manipulation of experimental parameters. In other words, experimental parameters that reduce the engagement of one memory system may as a consequence enhance the function of another system (Packard and Goodman, 2013). In the present study distributed training may have yielded poorer engagement of the hippocampal-dependent memory system (as evidenced by slow learning in the place task), which in turn may have freed the striatal system from competition in the response task and thus facilitated acquisition. The finding that place and response learning benefit from temporally distinct reinforcement parameters concords with the more general hypothesis that the hippocampus and DLS subserve relatively rapid and slow forms of learning, respectively. As mentioned above, the relative use of multiple memory systems can be influenced by amount of training. In dual-solution tasks that can be solved adequately with either memory system, limited training is associated with the preferential use of hippocampus-dependent spatial learning, whereas extended training is associated with DLSdependent response learning (Ritchie et al., 1950; Hicks, 1964; Packard and McGaugh, 1996; Packard, 1999). Similarly, extended training is associated with greater DLS-dependent habitual behavior in instrumental learning (Yin et al., 2006). Although the exact neurobiological mechanisms underlying this shift to striatumdependent memory remain undetermined, some evidence suggests a role for the neurotransmitter acetylcholine (Gold, 2004). For instance, over the course of training in a one-day dual-solution plus-maze task, acetylcholine release in the hippocampus peaks early in training, whereas acetylcholine release in the DLS increases slowly, peaking after about 60 trials or, in temporal terms, between approximately 60–100 min (Chang and Gold, 2003). This gradual rise in striatal acetylcholine correlates with the expression of response learning in the dual-solution plus-maze (Chang and Gold, 2003). Notably spaced training in the present study occurred for about 150 mins each day, which might allow for striatal acetylcholine release to reach its peak and promote response learning. In addition, in view of the present findings, the question arises whether the gradual learning subserved by the DLS may be a function of number of trials, passage of time, or both. It may be speculated that the mere “passage of time” allotted by either a spaced training protocol or “extended training” protocol with many trials may be sufficient to allow for the gradual rise in striatal acetylcholine, as well as other temporally graded neurobiological processes (Gold et al., 2013), that may facilitate response learning and/or impair place learning. It is also possible that these factors are not mutually exclusive and that an interaction between number of
trials and passage of time may govern the relative dominance of multiple memory systems. The present study suggests that memory systems benefit from temporally distinct reinforcement parameters, with spaced training selectively benefitting DLS-dependent memory and massed training benefitting hippocampus-dependent memory. Reinforcement parameters are but one class of experimental factors that may reveal differences between memory systems, whereas other important factors include stress/anxiety and the visual aspects of the learning environment (for review, see Packard and Goodman, 2013). Continued investigation of such experimental factors may not only aid in designing optimal conditions for studying multiple memory systems, but also may elucidate the characteristics of individual systems as well as the interactions that exist between them. References Abrams, M.L., Grice, J.K., 1976. Effects of practice and positional variables in acquisition of a complex psychomotor skill. Percept. Mot. Skills 43, 203–211. Albuoy, G., King, B.R., Maquet, P., Doyon, J., 2013. Hippocampus and striatum: dynamics and interaction during acquisition and sleep-related motor sequence memory consolidation. Hippocampus 23, 985–1004. Bilodeau, I.M., Schlosberg, H., 1951. Similarity in stimulating conditions as a variable in retroactive inhibition. J. Exp. Psychol. 41, 199–204. Bolhius, J.J., Bureˇsová, O., Bureˇs, J., 1985. Persistence of working memory of rats in an aversively motivated radial maze task. Behav. Brain Res. 15, 43–49. Bureˇsová, O., 1980. Spatial memory and instrumental learning. Acta Neurobiol. Exp. 40, 51–65. Chang, Q., Gold, P.E., 2003. Switching memory systems during learning: changes in patterns of brain acetylcholine release in the hippocampus and striatum in rats. J. Neurosci. 23, 3001–3005. Cohen, J., Reid, S., Chew, K., 1994. Effects of varying trial distribution, intra- and extramaze cues, and amount of reward on proactive interference in the radial maze. Anim. Learn. Behav. 22, 134–142. Commins, S., Cunningham, L., Harvey, D., Walsh, D., 2003. Massed but not spaced training impairs spatial memory. Behav. Brain Res. 139, 215–223. Domjan, M., 1980. Effects of the intertrial interval on taste-aversion learning in rats. Physiol. Behav. 25, 117–125. Ebbinghaus, H., 1913. Memory: A Contribution To Experimental Psychology. Teachers College, Columbia University, New York (Chapter 7). Glenberg, A.M., 1992. Distributed practice effects. In: Squire, L.R. (Ed.), Encyclopedia of Learning and Memory. MacMillan Publishing, New York. Gold, P.E., 2004. Coordination of multiple memory systems. Neurobiol. Learn. Mem. 82, 230–242. Gold, P.E., Newman, L.A., Scavuzzo, C.J., Korol, D.L., 2013. Modulation of multiple memory systems: from neurotransmitters to metabolic substrates. Hippocampus 23, 1053–1065. Goodrick, C.L., 1973. Maze learning of mature-young and aged rats as a function of distributed practice. J. Exp. Psychol. 98, 344–349. Heathers, G.L., 1940. The avoidance of repetition of a maze reaction in the rat as a function of the time interval between trials. J. Psychol. 10, 359–380. Hicks, L.H., 1964. Effects of overtraining on acquisition and reversal of place and response learning. Psychol. Rep. 15, 459–462. Honey, R.C., 1996. The temporal dynamics of a visual discrimination: the role of stimulus comparison and opponent processes. J. Exp. Psychol. Anim. B 22, 461–471. Hull, C.L., 1943. Principles Of Behavior. Appleton-Century, New York. Kesner, R.P., Bolland, B.L., Dakis, M., 1993. Memory for spatial locations, motor-responses, and objects -triple dissociation among the hippocampus, caudate-nucleus, and extrastriate visual-cortex. Exp. Brain Res. 93, 462–470. Kraemer, P.J., Randall, C.K., 1995. Spatial-learning in preweanling rats trained in a morris water maze. Psychobiology 23, 144–152. McBride, D.K., Payne, R.B., 1979. Reactive inhibition and intertrial correlations. J. Mot. Behav. 11, 253–259. McDonald, R.J., White, N.M., 1993. A triple dissociation of memory systems: hippocampus, amygdala, and dorsal striatum. Behav. Neurosci. 107, 3–22. McGeoch, J.A., 1932. Forgetting and the law of disuse. Psychol. Rev. 39, 352–370. Morris, R.G.M., Doyle, J., 1985. Successive incompatible tasks: evidence for separate subsystems for storage of spatial knowledge. In: Buzsaki, G., Vanderwolf, C.H. (Eds.), Electrical Activity of the Archicortex. Akademiai Kiado, Budapest. O’Brien, J.H., Packham, S.C., 1973. Conditioned leg movement in the cat with massed trials, trace conditioning, and weak us intensity. Cond. Reflex 8, 116–124. Olton, D.S., Becker, J.T., Handelman, G.E., 1979. Hippocampus, space, and memory. Behav. Brain Sci. 2, 313–322. Packard, M.G., 1999. Glutamate infused posttraining into the hippocampus or caudate-putamen differentially strengthens place and response learning. Proc.
Please cite this article in press as: Wingard, J.C., et al., Differential effects of massed and spaced training on place and response learning: A memory systems perspective. Behav. Process. (2015), http://dx.doi.org/10.1016/j.beproc.2015.06.004
319 320 321 322 323 324 325 326 327 328 329 330 331 332 333
334
335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394
G Model BEPROC 3067 1–5
ARTICLE IN PRESS J.C. Wingard et al. / Behavioural Processes xxx (2015) xxx–xxx
395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424
Natl. Acad. Sci. U. S. A. 96, 12881–12886. Packard, M.G., 2001. On the neurobiology of multiple memory systems: tolman versus hull, system interactions, and the emotion-memory link. Cogn. Process. 2, 3–24. Packard, M.G., 2009. Anxiety, cognition, and habit: a multiple memory systems perspective. Brain Res. 1293, 121–183. Packard, M.G., Goodman, J., 2012. Emotional arousal and multiple memory systems in the mammalian brain. Front. Behav. Neurosci. 6, 1–9. Packard, M.G., Goodman, J., 2013. Factors that influence the relative use of multiple memory systems. Hippocampus 23, 1044–1052. Packard, M.G., Hirsh, R., White, N.M., 1989. Differential effects of fornix and dorsolateral striatum lesions on two radial maze tasks: evidence for multiple memory systems. J. Neurosci. 9, 1465–1472. Packard, M.G., McGaugh, J.L., 1992. Double dissociation of fornix and caudate nucleus lesions on acquisition of two water maze tasks: further evidence for multiple memory systems. Behav. Neurosci. 106, 439–446. Packard, M.G., McGaugh, J.L., 1996. Inactivation of hippocampus or dorsolateral striatum with lidocaine differentially affects expression of place and response learning. Neurobiol. Learn. Mem. 65, 65–72. Panakhova, E., Bureˇsová, O., Bureˇs, J., 1984. Persistence of spatial memory in the Morris water tank task. Int. J. Psychophysiol. 2, 5–10. Poldrack, R.A., Packard, M.G., 2003. Competition among multiple memory systems: converging evidence from animal and human brain studies. Neuropsychologia 41, 245–251. Ritchie, B.F., Aeschliman, B., Pierce, P., 1950. Studies in spatial learning. VIII. Place performance and the acquisition of place dispositions. J. Comp. Physiol. Psychol. 43, 73–85. Roberts, W.A., Grant, D.S., 1976. Studies of short-term memory in the pigeon using the delayed matching to sample procedure. In: Medin, D.L., Roberts, W.A., Davis, R.T. (Eds.), Processes of Animal Memory. Lawrence Eribaum, Hillsdale NJ.
5
Salafia, W.R., Mis, F.W., Terry, W.S., Bartosia, R.S., Daston, A.P., 1973. Conditioning of nictitating-membrane response of rabbit (Oryctolagus-cuniculus) as a function of length and degree of variation of intertrial interval. Anim. Learn. Behav. 1, 109–115. Schroeder, J.A., Wingard, J., Packard, M.G., 2002. Post-training reversible inactivation of the dorsal hippocampus reveals interference between multiple memory systems. Hippocampus 12, 280–284. Solomon, R.L., 1948. The influence of work on behavior. Psychol. Bull. 45, 1–40. Spreng, M., Rossier, J., Schenk, F., 2002. Spaced training facilitates long-term retention of place navigation in adult but not in adolescent rats. Behav. Brain Res. 128, 103–108. Thompson, M.E., Thompson, J.P., 1949. Reactive inhibition as a factor in maze learning: ii. the role of reactive inhibition in studies of place learning versus response learning. J. Exp. Psychol. 39, 883–891. Underwood, B.J., 1957. Interference and forgetting. Psychol. Rev. 64, 49–60. White, N.M., McDonald, R.J., 2002. Multiple parallel memory systems in the brain of the rat. Neurobiol. Learn. Mem. 77, 125–184. White, N.M., Packard, M.G., McDonald, R.J., 2013. Dissociation of memory systems: the story unfolds. Behav. Neurosci. 127, 813–834. Wilson, M.A., McNaughton, B.L., 1994. Reactivation of hippocampal ensemble memories during sleep. Science 265, 676–679. Yin, H.H., Knowlton, B.J., 2004. Contributions of striatal subregions to place and response learning. Learn. Mem. 11, 459–463. Yin, H.H., Knowlton, B.J., Balleine, B.W., 2006. Inactivation of dorsolateral striatum enhances sensitivity to changes in the action-outcome contingency in instrumental conditioning. Behav. Brain Res. 166, 189–196. Zeaman, D., House, B.J., 1951. The growth and decay of reactive inhibition as measured by alternation behavior. J. Exp. Psychol. 41, 177–186.
Please cite this article in press as: Wingard, J.C., et al., Differential effects of massed and spaced training on place and response learning: A memory systems perspective. Behav. Process. (2015), http://dx.doi.org/10.1016/j.beproc.2015.06.004
425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452
ARTICLE IN PRESS
BEPROC 3067 1–5
Behavioural Processes xxx (2015) xxx–xxx
Contents lists available at ScienceDirect
Behavioural Processes journal homepage: www.elsevier.com/locate/behavproc
1
Short report
3
Differential effects of massed and spaced training on place and response learning: A memory systems perspective
4
Jeffrey C. Wingard b , Jarid Goodman a , Kah-Chung Leong a , Mark G. Packard a,b,∗
2
5 6
a b
Department of Psychology and Institute for Neuroscience, Texas A&M University, USA Department of Psychology, Yale University, USA
7
8 15
a r t i c l e
i n f o
a b s t r a c t
9
14
Article history: Received 31 March 2015 Received in revised form 8 May 2015 Accepted 1 June 2015 Available online xxx
16
1. Introduction
10 11 12 13
Studies employing brain lesion or intracerebral drug infusions in rats have demonstrated a double dissociation between the roles of the hippocampus and dorsolateral striatum in place and response learning. The hippocampus mediates a rapid cognitive learning process underlying place learning, whereas the dorsolateral striatum mediates a relatively slower learning process in which stimulus-response habits underlying response learning are acquired in an incremental fashion. One potential implication of these findings is that hippocampus-dependent learning may benefit from a relative massing of training trials, whereas dorsal striatum-dependent learning may benefit from a relative distribution of training trials. In order to examine this hypothesis, the present study compared the effects of massed (30 s inter-trial interval; ITI) or spaced (30 min ITI) training on acquisition of a hippocampus-dependent place learning task, and a dorsolateral striatum-dependent response task in a plus-maze. In the place task rats swam from varying start points (N or S) to a hidden escape platform located in a consistent spatial location (W). In the response task rats swam from varying start points (N or S) to a hidden escape platform located in the maze arm consistent with a body-turn response (left). In the place task, rats trained with the massed trial schedule acquired the task quicker than rats trained with the spaced trial schedule. In the response task, rats trained with the spaced trial schedule acquired the task quicker than rats trained with the massed trial schedule. The double dissociation observed suggests that the reinforcement parameters most conducive to effective learning in hippocampus-dependent and dorsolateral striatum-dependent learning may have differential temporal characteristics. © 2015 Published by Elsevier B.V.
Beneficial effects of spaced training over massed training schedules have been observed in various learning tasks across a range 18 of species. In lower animals this phenomenon has been reliably 19 20 demonstrated in various Pavlovian conditioning paradigms, includ21 ing for example eyelid conditioning (e.g., Salafia et al., 1973), taste 22 aversion conditioning (Domjan, 1980), and conditioned leg move23 ments (O’Brien and Packham, 1973). However, rare exceptions to 24 the facilitative effects of spaced training on learning do exist. In 25 rats, for instance, massed training results in superior acquisition of 26 “place” learning in both an appetitive plus-maze task (Thompson 27 and Thompson, 1949) and visuospatial discrimination task (Honey, 28Q1 1996). 17
∗ Corresponding author at: Department of Psychology, Texas A&M University, College Station, TX 77843, USA. Tel.: +1 979 845 9504; fax: +1 979 845 4727. E-mail address: [email protected] (M.G. Packard).
Such exceptions may potentially be explained within the context of multiple memory systems. According to this view, mammalian memory is organized in relatively independent brain systems that mediate different types of memory (for reviews see Packard, 2001; White and McDonald, 2002; Packard and Goodman, 2012; White et al., 2013). Several experiments investigating the neuroanatomical bases of multiple memory systems have revealed dissociable mnemonic roles for the hippocampus and dorsolateral striatum (DLS), in “cognitive” and stimulus-response “habit” memory, respectively (e.g., Packard et al., 1989; Packard and McGaugh, 1992; McDonald and White, 1993; Kesner et al., 1993). In addition to differing in their information content, the types of memory mediated by the hippocampus and DLS may differ in their temporal characteristics. The hippocampus is hypothesized to mediate a rapid form of cognitive learning, whereas the DLS mediates a habitual form of learning in which stimulus-response habits accrue in an incremental or gradual fashion. Thus, in “dual-solution” tasks in which both systems can provide an adequate learned solution, the relative contribution of the hippocampus and DLS changes over
http://dx.doi.org/10.1016/j.beproc.2015.06.004 0376-6357/© 2015 Published by Elsevier B.V.
Please cite this article in press as: Wingard, J.C., et al., Differential effects of massed and spaced training on place and response learning: A memory systems perspective. Behav. Process. (2015), http://dx.doi.org/10.1016/j.beproc.2015.06.004
29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47
G Model BEPROC 3067 1–5
J.C. Wingard et al. / Behavioural Processes xxx (2015) xxx–xxx
49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86
the course of training. Specifically, hippocampus-dependent cognitive learning dominates early in training, whereas DLS-dependent habit learning dominates after extended training (e.g., Packard and McGaugh, 1996). One implication of the idea that the hippocampus and DLS mediate rapid and slow forms of learning, respectively, is that the former memory system may benefit from a relative massing of training trials, whereas the latter system may benefit from a relative distribution of training trials. This is consistent with the observation that the DLS-dependent memory system typically comes “online” later in training than the hippocampus-dependent memory system (Ritchie et al., 1950; Hicks, 1964; Packard and McGaugh, 1996; Chang and Gold, 2003), potentially making spaced training more conducive to response learning. The relative effectiveness of massed and spaced training protocols in place and response learning may also be interpreted within the context of classical learning theories (Ebbinghaus, 1913; McGeoch, 1932; Hull, 1943; Roberts and Grant, 1976), as well as the observation that hippocampus- and DLS-dependent memory systems sometimes compete for control of learning (Poldrack and Packard, 2003). The present study examined whether massed or spaced training differentially benefit the forms of learning mediated by the hippocampus and DLS. Accordingly, rats received either massed training (i.e., short inter-trial interval) or distributed (i.e., long inter-trial interval) training in two water plus-maze tasks, one task that required hippocampus-dependent “place” learning and another task that required DLS-dependent “response” learning. Importantly, the place and response learning tasks included the same motivational, motoric, and sensory requirements and were also conducted in the same maze environment. These tasks only differed in terms of the learning requirement, and thus any difference between massed and spaced training protocols in the place and response tasks may be more likely attributed to the protocols differentially affecting place and response learning, as opposed to an effect on non-mnemonic processes. It was hypothesized that massed training would result in superior learning in the hippocampus-dependent place task, whereas spaced training would result in superior learning in the DLS-dependent response task.
87
2. General methods
88
2.1. Subjects
95
Subjects were 35 experimentally naïve male Charles River Long–Evans rats (weighing 300–350 g at arrival), individually housed in a climate-controlled vivarium with food and water provided ad libitum. After arrival, animals were given a one-week acclimation period before experiments began. Animals were on a 12:12-h light:dark cycle (lights on at 7 a.m.), and experiments were conducted during the light phase of this cycle.
96
2.2. Apparatus
89 90 91 92 93 94
97 98 99 100 101 102 103 104 105 106 107
Rats were trained in a black circular water maze (1.83 m diameter, 0.58 m in height) into which a clear Plexiglas plusmaze (59.69 cm height, arm-width of 25.4 cm, and arm-length of 43.18 cm) was inserted. The maze was filled to a water level of 20 cm. The water temperature was held constant at 25 ◦ C throughout training. Extra-maze visual cues of various geometric shapes were placed on square dark green curtains surrounding the water maze (none of the extra-maze cues were placed in a spatially congruent/proximal position with the ends of the maze arms). An invisible black Plexiglas escape platform (11 × 14 × 19 cm) was located in one arm of the maze and submerged 1 cm below water
Place Learning 100 90
Percent Correct
2 48
ARTICLE IN PRESS
80 70 60 50
Spaced ITI
40
Massed ITI
30 20 0
1
2
3
4
5
Day Fig. 1. Effect of training with a spaced or massed inter-trial interval (ITI) on the acquisition of a single-solution place water plus-maze task. Relative to animals given spaced training (30 min ITIs), animals given massed training (30 s ITIs) displayed a greater percentage of correct arm entries over the course of training in the place learning task.
level. Swimming behavior was observed and recorded by an experimenter standing outside the curtains and using an overhead video camera system and monitor. 2.3. Behavioral procedures Behavioral training was conducted over 5 consecutive days, and each day consisted of 6 trials (i.e. swims) in either a place or response plus-maze task. An inter-trial interval (ITI) of 30 min was used for rats given spaced training, and an ITI of 30 s was used for rats given massed training. The rats started from either the north or south arms. On odd days, the sequence of start positions was NSSNNS; and on even days, SNNSSN. On a given trial, a clear Plexiglas barrier blocked access to the arm directly opposite the start arm. For the place task, rats (spaced training, n = 7; massed training, n = 7) were trained to swim to a submerged escape platform that was always located in the same goal arm of the maze (west). In the response task (spaced training, n = 11; massed training, n = 10), the escape platform was always located in the maze goal arm to the left of the start arm, requiring the rat to learn a specific body-turn response to approach the escape platform (i.e. when the start arm was north, the platform was in the west arm; and when the start arm was south, the platform was in the east arm). After reaching the escape platform, rats remained on it for 10 s. Rats that failed to find the hidden platform within 60 s were manually guided to it. The experimenter recorded the number of errors made on each trial. An error was defined as swimming a full body length into the non-goal maze arm. After removal from the escape platform, rats were returned to their cages located on shelving adjacent to and outside of the behavioral testing room for the assigned ITI. After a rat finished a training session, the rat was again returned to its holding cage until all rats were finished training for the day. In this way, animals in the spaced and massed conditions spent an equal amount of time in the holding cage on the table adjacent to the maze room for each day of training. 3. Results Fig. 1 depicts the mean percentage of correct responses during training in the place task. A two-way one-repeated measure ANOVA (group × training day) computed on percent correct data for all five days revealed a significant main effect of group [F (1, 12) = 7.1, p < 0.05]. Main effect of training day [F (4, 48) = 2.4, n.s.],
Please cite this article in press as: Wingard, J.C., et al., Differential effects of massed and spaced training on place and response learning: A memory systems perspective. Behav. Process. (2015), http://dx.doi.org/10.1016/j.beproc.2015.06.004
108 109 110
111
112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140
141
142 143 144 145 146
G Model
ARTICLE IN PRESS
BEPROC 3067 1–5
J.C. Wingard et al. / Behavioural Processes xxx (2015) xxx–xxx
Response Learning 100
Percent Correct
90 80 70 60 50
Spaced ITI
40
Massed ITI
30 20 0
1
2
3
4
5
Day Fig. 2. Effect of training with a spaced or massed inter-trial interval (ITI) on acquisition of a single-solution response water plus-maze task. Relative to animals given massed training (30 s ITIs), animals given spaced training (30 min ITIs) displayed a greater percentage of correct arm entries over the course of training in the response learning task.
159
and the training day × group interaction [F (4, 48) = 0.12, n.s.] were not significant. These data indicate that in the place learning task rats receiving massed training displayed superior learning relative to rats receiving spaced training. Fig. 2 depicts the mean percentage of correct responses during training in the response task. A two-way one-repeated measure ANOVA (group × training day) computed on percent correct data for all five days revealed a main effect of group [F (1, 9) = 4.6, p < 0.05], and training day [F (4, 76) = 14.9, p < 0.001]. The group × training day interaction was not significant [F (4, 76) = 0.67, n.s.]. These data indicate that in the response learning task rats receiving spaced training displayed superior learning relative to the rats receiving massed training.
160
4. Discussion
147 148 149 150 151 152 153 154 155 156 157 158
161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187
The findings indicate a double dissociation of the effects of massed and spaced training on place and response learning. Specifically, that massed training, relative to spaced training, enhances acquisition in a “place” plus-maze task, whereas spaced training enhances acquisition in the “response” task. These findings are consistent with a similar study that used appetitive versions of the place and response plus-maze tasks (Thompson and Thompson, 1949). In view of extensive evidence from neurobehavioral studies indicating that place and response learning in the plus-maze critically depend on the hippocampus and DLS, respectively (Packard and McGaugh, 1996; Packard, 1999; Chang and Gold, 2003; Yin and Knowlton, 2004), massed training may in general benefit learning mediated by the hippocampus, whereas spaced training may benefit learning mediated by the DLS. Given the aversive nature of the water plus-maze tasks and the role that stress/anxiety plays in multiple memory systems (Packard, 2009), it should be noted that stress/anxiety might have contributed to the effects of massed and spaced training protocols in the present study. However previous findings indicating that similar effects may be observed in appetitive versions of the place and response plus-maze tasks (Thompson and Thompson, 1949) suggest that the effects of massed and spaced training protocols may persist regardless of the stressful nature of the learning task. The enhancing effect of spaced training on response learning is consistent with a large body of evidence showing distributed practice improves learning in motor/procedural-like tasks in lower animals (e.g. Salafia et al., 1973; Domjan, 1980) and humans (e.g.,
3
Abrams and Grice, 1976; McBride and Payne, 1979; Glenberg, 1992). On the other hand, the finding that a relative massing of trials benefits spatial learning appears at first glance inconsistent with previous research, which has shown in contrast a benefit of spaced training in various spatial tasks (e.g., Goodrick, 1973; Cohen et al., 1994), including spatial learning in the water maze (Morris and Doyle, 1985; Kraemer and Randall, 1995; Spreng et al., 2002; Commins et al., 2003). One possible reason for this inconsistency is that previous studies used different parameters to define massed and spaced training conditions. Massed training in the water maze, for instance, has typically been conducted over 1 day, whereas spaced training has consisted of the same number of trials, but with the trials distributed over multiple days (e.g., Morris and Doyle, 2003). Thus, spaced training in such previous studies may have permitted long-term consolidation processes (like those that purportedly occur during sleep; Wilson and McNaughton, 1994; Albuoy et al., 2013), which otherwise may have been prevented by a 1-day massed training protocol. The design of the present study avoided this potential confound by ensuring that the number of training days was equal between massed and spaced conditions, whereas the ITI within a daily training session differed between the groups. In this way, the relative impairing effect of spaced training observed in the present spatial task is consistent with other previous work showing that longer ITIs impair acquisition and retention in a variety of spatial tasks (Thompson and Thompson, 1949; Olton et al., 1979; Bureˇsová, 1980; Panakhova et al., 1984; Bolhius et al., 1985). With regard to the mechanism(s) that may underlie the present findings, several principles in classical learning theory may be proposed, including the trace decay hypothesis (Ebbinghaus, 1913; Roberts and Grant, 1976), interference theory (McGeoch, 1932), and reactive inhibition (Hull, 1943). The trace decay hypothesis suggests that after an event is terminated, the memory trace for that event begins to deteriorate over time. In the Morris water maze, increasing the intertrial interval between an initial training trial and subsequent test trial is associated with greater escape latencies (Panakhova et al., 1984). In the present study, rats receiving distributed training in the spatial task may have experienced greater memory decay over the 30-min intertrial intervals, relative to the 30-s intervals, and for this reason displayed slower acquisition. However, trace decay fails to account for the beneficial effect of distributed training in the response task. In this case, findings in the response task may be partly explained by interference theory. Interference putatively occurs when one memory trace interferes with the retrieval of another trace (McGeoch, 1932), and this interference becomes more likely when the two traces are acquired in the same environment (Bilodeau and Schlosberg, 1951; Underwood, 1957). During acquisition of the response task, rats may in the beginning encode both the spatial location of the platform in a given trial and also the turning response used to reach the platform (consistent with the hypothesis of “multiple parallel memory systems”; White and McDonald, 2002). However, over the course of training in the response task, memory of the platform’s spatial location in a previous trial may interfere with the retrieval of a turning response that sometimes leads to a different spatial location. This would result in occasional errors. Nevertheless, considering that interference from a previous memory may dissipate over time (Underwood, 1957), longer ITIs could potentially prevent interference from the old spatial memory, permitting uninhibited retrieval of the turning response and, thus, fewer errors over the course of training in the response task. Another mechanism that may underlie the benefit of distributed training in the “response” plus-maze task is reactive inhibition (Hull, 1943). Reactive inhibition refers to a hypothesized “negative drive” that an organism experiences after performing a rewarded response; this negative drive thus makes the animal less likely
Please cite this article in press as: Wingard, J.C., et al., Differential effects of massed and spaced training on place and response learning: A memory systems perspective. Behav. Process. (2015), http://dx.doi.org/10.1016/j.beproc.2015.06.004
188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253
G Model BEPROC 3067 1–5 4 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318
ARTICLE IN PRESS J.C. Wingard et al. / Behavioural Processes xxx (2015) xxx–xxx
to repeat that same response for a brief time. A purported consequence of reactive inhibition is the tendency for an animal to spontaneously alternate choice behavior in the T-maze (Solomon, 1948), and in this way reactive inhibition may lead to incorrect turns in learning situations that require the same turning response across trials (Thompson and Thompson, 1949), as in the present “response” plus-maze task. However, as reactive inhibition may gradually diminish over time (Hull, 1943), longer intertrial intervals could reduce this tendency to alternate (Heathers, 1940; Zeaman and House, 1951) and therefore, in the present study, could have prevented incorrect turns in the response task. An additional consideration is that the memory systems mediating place and response learning in these tasks can sometimes “compete” for control of learning (Poldrack and Packard, 2003). An implication of this competitive interaction is that by impairing one memory system, learning that is mediated by the other “intact” system may be facilitated. For instance, lesions of the hippocampal system facilitate acquisition in a variety of dorsal striatumdependent response tasks (Packard et al., 1989; McDonald and White, 1993; Schroeder et al., 2002). A similar process may emerge with the mere manipulation of experimental parameters. In other words, experimental parameters that reduce the engagement of one memory system may as a consequence enhance the function of another system (Packard and Goodman, 2013). In the present study distributed training may have yielded poorer engagement of the hippocampal-dependent memory system (as evidenced by slow learning in the place task), which in turn may have freed the striatal system from competition in the response task and thus facilitated acquisition. The finding that place and response learning benefit from temporally distinct reinforcement parameters concords with the more general hypothesis that the hippocampus and DLS subserve relatively rapid and slow forms of learning, respectively. As mentioned above, the relative use of multiple memory systems can be influenced by amount of training. In dual-solution tasks that can be solved adequately with either memory system, limited training is associated with the preferential use of hippocampus-dependent spatial learning, whereas extended training is associated with DLSdependent response learning (Ritchie et al., 1950; Hicks, 1964; Packard and McGaugh, 1996; Packard, 1999). Similarly, extended training is associated with greater DLS-dependent habitual behavior in instrumental learning (Yin et al., 2006). Although the exact neurobiological mechanisms underlying this shift to striatumdependent memory remain undetermined, some evidence suggests a role for the neurotransmitter acetylcholine (Gold, 2004). For instance, over the course of training in a one-day dual-solution plus-maze task, acetylcholine release in the hippocampus peaks early in training, whereas acetylcholine release in the DLS increases slowly, peaking after about 60 trials or, in temporal terms, between approximately 60–100 min (Chang and Gold, 2003). This gradual rise in striatal acetylcholine correlates with the expression of response learning in the dual-solution plus-maze (Chang and Gold, 2003). Notably spaced training in the present study occurred for about 150 mins each day, which might allow for striatal acetylcholine release to reach its peak and promote response learning. In addition, in view of the present findings, the question arises whether the gradual learning subserved by the DLS may be a function of number of trials, passage of time, or both. It may be speculated that the mere “passage of time” allotted by either a spaced training protocol or “extended training” protocol with many trials may be sufficient to allow for the gradual rise in striatal acetylcholine, as well as other temporally graded neurobiological processes (Gold et al., 2013), that may facilitate response learning and/or impair place learning. It is also possible that these factors are not mutually exclusive and that an interaction between number of
trials and passage of time may govern the relative dominance of multiple memory systems. The present study suggests that memory systems benefit from temporally distinct reinforcement parameters, with spaced training selectively benefitting DLS-dependent memory and massed training benefitting hippocampus-dependent memory. Reinforcement parameters are but one class of experimental factors that may reveal differences between memory systems, whereas other important factors include stress/anxiety and the visual aspects of the learning environment (for review, see Packard and Goodman, 2013). Continued investigation of such experimental factors may not only aid in designing optimal conditions for studying multiple memory systems, but also may elucidate the characteristics of individual systems as well as the interactions that exist between them. References Abrams, M.L., Grice, J.K., 1976. Effects of practice and positional variables in acquisition of a complex psychomotor skill. Percept. Mot. Skills 43, 203–211. Albuoy, G., King, B.R., Maquet, P., Doyon, J., 2013. Hippocampus and striatum: dynamics and interaction during acquisition and sleep-related motor sequence memory consolidation. Hippocampus 23, 985–1004. Bilodeau, I.M., Schlosberg, H., 1951. Similarity in stimulating conditions as a variable in retroactive inhibition. J. Exp. Psychol. 41, 199–204. Bolhius, J.J., Bureˇsová, O., Bureˇs, J., 1985. Persistence of working memory of rats in an aversively motivated radial maze task. Behav. Brain Res. 15, 43–49. Bureˇsová, O., 1980. Spatial memory and instrumental learning. Acta Neurobiol. Exp. 40, 51–65. Chang, Q., Gold, P.E., 2003. Switching memory systems during learning: changes in patterns of brain acetylcholine release in the hippocampus and striatum in rats. J. Neurosci. 23, 3001–3005. Cohen, J., Reid, S., Chew, K., 1994. Effects of varying trial distribution, intra- and extramaze cues, and amount of reward on proactive interference in the radial maze. Anim. Learn. Behav. 22, 134–142. Commins, S., Cunningham, L., Harvey, D., Walsh, D., 2003. Massed but not spaced training impairs spatial memory. Behav. Brain Res. 139, 215–223. Domjan, M., 1980. Effects of the intertrial interval on taste-aversion learning in rats. Physiol. Behav. 25, 117–125. Ebbinghaus, H., 1913. Memory: A Contribution To Experimental Psychology. Teachers College, Columbia University, New York (Chapter 7). Glenberg, A.M., 1992. Distributed practice effects. In: Squire, L.R. (Ed.), Encyclopedia of Learning and Memory. MacMillan Publishing, New York. Gold, P.E., 2004. Coordination of multiple memory systems. Neurobiol. Learn. Mem. 82, 230–242. Gold, P.E., Newman, L.A., Scavuzzo, C.J., Korol, D.L., 2013. Modulation of multiple memory systems: from neurotransmitters to metabolic substrates. Hippocampus 23, 1053–1065. Goodrick, C.L., 1973. Maze learning of mature-young and aged rats as a function of distributed practice. J. Exp. Psychol. 98, 344–349. Heathers, G.L., 1940. The avoidance of repetition of a maze reaction in the rat as a function of the time interval between trials. J. Psychol. 10, 359–380. Hicks, L.H., 1964. Effects of overtraining on acquisition and reversal of place and response learning. Psychol. Rep. 15, 459–462. Honey, R.C., 1996. The temporal dynamics of a visual discrimination: the role of stimulus comparison and opponent processes. J. Exp. Psychol. Anim. B 22, 461–471. Hull, C.L., 1943. Principles Of Behavior. Appleton-Century, New York. Kesner, R.P., Bolland, B.L., Dakis, M., 1993. Memory for spatial locations, motor-responses, and objects -triple dissociation among the hippocampus, caudate-nucleus, and extrastriate visual-cortex. Exp. Brain Res. 93, 462–470. Kraemer, P.J., Randall, C.K., 1995. Spatial-learning in preweanling rats trained in a morris water maze. Psychobiology 23, 144–152. McBride, D.K., Payne, R.B., 1979. Reactive inhibition and intertrial correlations. J. Mot. Behav. 11, 253–259. McDonald, R.J., White, N.M., 1993. A triple dissociation of memory systems: hippocampus, amygdala, and dorsal striatum. Behav. Neurosci. 107, 3–22. McGeoch, J.A., 1932. Forgetting and the law of disuse. Psychol. Rev. 39, 352–370. Morris, R.G.M., Doyle, J., 1985. Successive incompatible tasks: evidence for separate subsystems for storage of spatial knowledge. In: Buzsaki, G., Vanderwolf, C.H. (Eds.), Electrical Activity of the Archicortex. Akademiai Kiado, Budapest. O’Brien, J.H., Packham, S.C., 1973. Conditioned leg movement in the cat with massed trials, trace conditioning, and weak us intensity. Cond. Reflex 8, 116–124. Olton, D.S., Becker, J.T., Handelman, G.E., 1979. Hippocampus, space, and memory. Behav. Brain Sci. 2, 313–322. Packard, M.G., 1999. Glutamate infused posttraining into the hippocampus or caudate-putamen differentially strengthens place and response learning. Proc.
Please cite this article in press as: Wingard, J.C., et al., Differential effects of massed and spaced training on place and response learning: A memory systems perspective. Behav. Process. (2015), http://dx.doi.org/10.1016/j.beproc.2015.06.004
319 320 321 322 323 324 325 326 327 328 329 330 331 332 333
334
335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394
G Model BEPROC 3067 1–5
ARTICLE IN PRESS J.C. Wingard et al. / Behavioural Processes xxx (2015) xxx–xxx
395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424
Natl. Acad. Sci. U. S. A. 96, 12881–12886. Packard, M.G., 2001. On the neurobiology of multiple memory systems: tolman versus hull, system interactions, and the emotion-memory link. Cogn. Process. 2, 3–24. Packard, M.G., 2009. Anxiety, cognition, and habit: a multiple memory systems perspective. Brain Res. 1293, 121–183. Packard, M.G., Goodman, J., 2012. Emotional arousal and multiple memory systems in the mammalian brain. Front. Behav. Neurosci. 6, 1–9. Packard, M.G., Goodman, J., 2013. Factors that influence the relative use of multiple memory systems. Hippocampus 23, 1044–1052. Packard, M.G., Hirsh, R., White, N.M., 1989. Differential effects of fornix and dorsolateral striatum lesions on two radial maze tasks: evidence for multiple memory systems. J. Neurosci. 9, 1465–1472. Packard, M.G., McGaugh, J.L., 1992. Double dissociation of fornix and caudate nucleus lesions on acquisition of two water maze tasks: further evidence for multiple memory systems. Behav. Neurosci. 106, 439–446. Packard, M.G., McGaugh, J.L., 1996. Inactivation of hippocampus or dorsolateral striatum with lidocaine differentially affects expression of place and response learning. Neurobiol. Learn. Mem. 65, 65–72. Panakhova, E., Bureˇsová, O., Bureˇs, J., 1984. Persistence of spatial memory in the Morris water tank task. Int. J. Psychophysiol. 2, 5–10. Poldrack, R.A., Packard, M.G., 2003. Competition among multiple memory systems: converging evidence from animal and human brain studies. Neuropsychologia 41, 245–251. Ritchie, B.F., Aeschliman, B., Pierce, P., 1950. Studies in spatial learning. VIII. Place performance and the acquisition of place dispositions. J. Comp. Physiol. Psychol. 43, 73–85. Roberts, W.A., Grant, D.S., 1976. Studies of short-term memory in the pigeon using the delayed matching to sample procedure. In: Medin, D.L., Roberts, W.A., Davis, R.T. (Eds.), Processes of Animal Memory. Lawrence Eribaum, Hillsdale NJ.
5
Salafia, W.R., Mis, F.W., Terry, W.S., Bartosia, R.S., Daston, A.P., 1973. Conditioning of nictitating-membrane response of rabbit (Oryctolagus-cuniculus) as a function of length and degree of variation of intertrial interval. Anim. Learn. Behav. 1, 109–115. Schroeder, J.A., Wingard, J., Packard, M.G., 2002. Post-training reversible inactivation of the dorsal hippocampus reveals interference between multiple memory systems. Hippocampus 12, 280–284. Solomon, R.L., 1948. The influence of work on behavior. Psychol. Bull. 45, 1–40. Spreng, M., Rossier, J., Schenk, F., 2002. Spaced training facilitates long-term retention of place navigation in adult but not in adolescent rats. Behav. Brain Res. 128, 103–108. Thompson, M.E., Thompson, J.P., 1949. Reactive inhibition as a factor in maze learning: ii. the role of reactive inhibition in studies of place learning versus response learning. J. Exp. Psychol. 39, 883–891. Underwood, B.J., 1957. Interference and forgetting. Psychol. Rev. 64, 49–60. White, N.M., McDonald, R.J., 2002. Multiple parallel memory systems in the brain of the rat. Neurobiol. Learn. Mem. 77, 125–184. White, N.M., Packard, M.G., McDonald, R.J., 2013. Dissociation of memory systems: the story unfolds. Behav. Neurosci. 127, 813–834. Wilson, M.A., McNaughton, B.L., 1994. Reactivation of hippocampal ensemble memories during sleep. Science 265, 676–679. Yin, H.H., Knowlton, B.J., 2004. Contributions of striatal subregions to place and response learning. Learn. Mem. 11, 459–463. Yin, H.H., Knowlton, B.J., Balleine, B.W., 2006. Inactivation of dorsolateral striatum enhances sensitivity to changes in the action-outcome contingency in instrumental conditioning. Behav. Brain Res. 166, 189–196. Zeaman, D., House, B.J., 1951. The growth and decay of reactive inhibition as measured by alternation behavior. J. Exp. Psychol. 41, 177–186.
Please cite this article in press as: Wingard, J.C., et al., Differential effects of massed and spaced training on place and response learning: A memory systems perspective. Behav. Process. (2015), http://dx.doi.org/10.1016/j.beproc.2015.06.004
425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452
