A Tactile-visual Conditional Discrimination Task for Testing Spatial Working Memory in Rats

引用 收藏 提问与回复 分享您的反馈 Cited by



The Journal of Neuroscience
Aug 2016



This protocol describes a novel dual task comparison across two variants of a tactile-visual conditional discrimination (CD) T-maze task, one is dependent upon spatial working memory (SWM; CDWM) and the other one (CDSTANDARD) is not. The task variants are equivalent in their sensory and motor requirements and overt behavior of the rat. Therefore, differences between the two task variants in the dependent variables such as choice accuracy, neural firing patterns, and the effects of pharmacological or optogenetic inactivation in brain regions of interest can be attributed to SWM, ruling out confounding sensorimotor variables, such as tactile, visual and self-motion cues. The CDWM task protocol is published in Hallock et al., 2013b and Urban et al., 2014.

Keywords: Spatial working memory (空间工作记忆), Conditional discrimination (条件辨别), T-maze (T形迷宫), Encoding (编码), Retrieval (回忆)


Our laboratory is interested in exploring the neural mechanisms of working memory. Therefore, we have developed a task that can be used to assess spatial working memory (SWM) ability in rats. Working memory is defined as holding a limited amount of information ‘online’ so that the information can be used or manipulated to guide goal-directed behavior (Baddley, 1992). Because rodents are naturally inclined to forage for food, they are excellent models to use to probe SWM. Our laboratory has developed and used a conditional discrimination (CD) T-maze task in which floor inserts that vary in texture and color serve as conditional cues for the rewarded goal arm (Griffin et al., 2012; Hallock and Griffin, 2013; Hallock et al., 2013a; Shaw et al., 2013; Hallock et al., 2016). For example, the rats learn to choose the left goal arm if they encounter a mesh insert and the right goal arm if they encounter a wooden insert. To discriminate between the inserts, rats can use visual information (black vs. light brown), tactile information (rough mesh vs. smooth wood), or a combination of both types of information. Because the insert covers the entire floor of the maze and is available when the rat makes a goal-arm choice, this task does not require SWM. More recently, we have developed a working-memory variant of the task (CDWM; Hallock et al., 2013b; Urban et al., 2014). In this variant of the task, the floor insert cues extend only halfway up the central arm of the maze and are not available when the rat makes a goal-arm choice, thus requiring the rats to hold the cue in mind for a brief period of time in order to make a correct choice and receive food reward. In an ongoing experiment, we have found that it is possible to train rats on both variants of the task, giving us a powerful way to identify behavioral correlates of SWM while ruling out confounding sensorimotor variables such as visual, tactile, and self-motion cues.

Materials and Reagents

  1. Male Long Evans Hooded (Harlan, Indianapolis) rats, weighing between 250 g and 500 g upon arrival (approximately 90 days of age)
  2. Chocolate Sprinkles are used for food reward. Our lab uses the Chef’s Quality brand
  3. 70% ethanol for cleaning the maze between daily training sessions


  1. Wooden T-Maze that consists of a central stem (117 x 10 x 5 cm), two goal arms (56.5 x 10 x 5 cm) and two return arms (112 x 10 x 5 cm). The floor of the maze is covered with Roppe black vinyl (3 mm thick) (Figure 1)
    Note: The T-maze was custom-built by members of our lab.

    Figure 1. T-Maze used for both variants of the CD task shown with (A) and without (B) the removable barrier is used to confine the rat to the start box during the intertrial interval

  2. A wooden stool (height: 69 cm) with a plastic saucer (diameter: 38 cm) attached to the seat serves, as the start box, is positioned at the base of the maze. The start box is separated from the maze by a removable 6-cm tall wooden barrier
  3. Three removable wooden floor inserts covered with black plastic mesh on one side and smooth wood on the other serve as conditional cues (Figure 2). The black mesh was glued to one side of the inserts with superglue, and consisted of 4 x 4 mm mesh squares. For the CDWM variant of the task, the central arm insert (74 x 8 cm) extends halfway from the start box to the T-intersection. For the CDSTANDARD variant of the task, the central arm insert (117 x 8 cm) covers the entire length of the central arm from the start box to the T-intersection. The three goal arm inserts, one large insert (61 x 8 cm) and two small inserts (26 x 8 cm) are placed at the ends of the goal arms next to the reward cups

    Figure 2. Removable wooden inserts shown mesh side up used for the CDWM (A) and CDSTANDARD (B) variants of the task, and close up view of the insert (C)

  4. Black curtain, surrounding entire behavior room, 51 cm away from the maze
  5. Distal cues taped to black curtain, 142 cm from the floor, located behind the start pedestal, left and right reward cups (Figure 3). The cues are a pink triangle (38.1 x 29.2 cm), a red X (40.6 x 35.6 cm) and a blue cube (43.2 x 41.9 cm) made out of colored tape

    Figure 3. Distal cues taped to the curtains that surround the maze above the reward cups (A) and above the start box (B)

  6. Plastic cups (3 cm diameter; 1 cm depth) located at the end of each goal arm where chocolate sprinkles (4-5 pieces) are delivered. The caps were 20 oz. water/soda bottle caps
  7. One 60 W incandescent lamp attached to the curtain track of the back middle wall, near the ceiling. The lamp is facing up towards the ceiling, therefore the room is dimly lit with no direct lighting on any portion of the maze


See Figure 4 for a flowchart of the experimental procedures.

Figure 4. Flowchart of the experimental procedures, starting with acclimation of the rats to the vivarium through dual-task training

  1. Rats are housed 2-3 per cage in a temperature and humidity-controlled animal vivarium and kept on a controlled light cycle (7:00 AM-7:00 PM). Rats are given ad libitum access to food (Prolab RMH 3000 rat chow pellets) and water during their acclimation to the colony room.
  2. After a 7-day acclimation period in the vivarium, rats are individually housed and brought to the laboratory daily for 5-7 days and handled for 15 min per day by experimenter. The experimenter handles the rat by placing a diaper over their lap and placing the animal on top of the diaper, allowing the animal to move freely. After handling, the experimenter places a plastic cup of chocolate sprinkles in the home cage. The rats are given 20 min to eat the sprinkles, and after 20 min the cup is removed. Rats are then put on food restriction (4-5 food pellets daily) to maintain them between 80 and 90% of their free-feeding body weight and given ad libitum access to water. Rats are food restricted starting at pre-training (i.e., the handling period) until the end of testing. Starting weight is recorded on the first day of handling and rats are weighed weekly for the duration of the experiment.
  3. Next, rats undergo goal box training. In this phase, rats are brought to the behavior room where the T-maze is located. The animals are confined to the goal arms on the maze and allowed to eat chocolate sprinkles from the reward cups for a total of 6 daily trials (3 per goal arm). A single trial has a duration of 90 sec with an ITI of 10 sec. The trial is terminated early if the animal consumes the food reward in less than 90 sec. Once rats consume the reward in less than 90 sec on every trial for two consecutive sessions, they progress to forced-run training. It typically takes 3 to 5 days for animals to reach goal box training performance criterion.
  4. Forced run sessions consist of 12 trials per day, with 6 left and 6 right trials given in a pseudorandom sequence. Prior to each trial, the experimenter places a removable wooden barrier at the entrance of one of the goal arms. At the start of the trial, the rat is confined to the start box with a second wooden removable barrier. The forced-run trial begins when the start box barrier is removed. The rat is encouraged to run from the start box, down the central arm, turn down the available goal arm, consume the reward, and return to the start box via the return arm. The rat is discouraged from turning around at any point on the maze by blocking his path with wooden stick. This path correction procedure is only necessary on early training sessions after which the experimenter interacts with the rat as minimally as possible. Once rats consume reward on 80% of forced run trials for 2 consecutive sessions, they progress to the single-task phase of training. Forced run training typically lasts for 6 to 7 days.
  5. Rats are trained on the CDWM task (Figure 5) or the CDSTANDARD task (Figure 6). The starting task variant is counterbalanced between rats, resulting in two groups; the CDWM group and the CDSTANDARD group.

    Figure 5. Schematic of CDWM task. Removable wooden insert with plastic black mesh on one side and smooth wood on the other, line the first half of the central arm. See Video 1.

    Video 1. Example of three trials (2 mesh trials followed by 1 wood trial) of the CDWM task. The experimenter sham-flips the insert between the two mesh trials and flips the insert between the mesh and wood trial. She then places the food reward in one of the food cups and places the baited cup at the end of the correct goal arm.

    Figure 6. Schematic of the CDSTANDARD task. Removable wooden inserts covered with plastic black mesh on one side and smooth wood on the other cover the entire central arm and goal arms. See Video 2.

    Video 2. Example of three trials (1 wood followed by 2 mesh trials) of the CDSTANDARD task. Similar to Video 1, the experimenter flips the inserts between the wood and mesh trials and sham-flips the inserts between the two mesh trials. She then places the food reward in one of the food cups and places the baited cup at the end of the correct goal arm.

    1. Prior to each trial, the experimenter places the floor inserts onto the central stem and both goal arms of the T-maze. The two short left and right goal arm inserts and the half sized central stem insert will be the same color/texture. The goal inserts are used to help the rat learn and maintain the association of the conditional discrimination cue and the location of the food reward (e.g., Wood-Right or Mesh-Left). Our reasoning is that the absence or presence of food will enhance learning of the conditional discrimination if it occurs concomitantly with the presentation of the conditional cue. In the CDWM task, the inserts are half the size of the CDSTANDARD inserts. The central arm insert covers the first half of the central stem of the maze. The goal arm inserts cover the last half of the goal arms, farthest from the T-intersection. One side of the maze insert is covered with black mesh, and the other side is smooth wood (light brown). We have found that it is not necessary to fasten the removable inserts to the maze. The inserts fit snugly in the maze, so they are placed in the same location on every trial.
    2. In both tasks, conditional cues are associated with the location of food reward on the T-maze Rats are required to select either the left or right goal based on the texture and color (smooth wood or black mesh) of the floor insert. For example, rats learn to turn right at the T-intersection of the maze when they experience smooth, wood and turn left when they experience black mesh. Trials are presented in a pseudorandom sequence with equal numbers of mesh and wood trials per session (Fellows, 1967). The reward contingency is counterbalanced across rats, with half of the rats in an experiment learning to select the right goal arm on a mesh trial and the left goal arm on a wood trial and the other half of the rats learning the opposite rule (left on mesh, right on wood). The cue inserts are flipped between each trial, even if the same cue is presented on consecutive trials. Flipping cue inserts is done in order to ensure that the rat cannot solve the task by using auditory cues.
    3. Between each trial, rats wait in the start box for 20 sec while the experimenter prepared the maze for the next trial. A typical training session consisting of 24 trials (12 wood 12 mesh) takes about 30-40 min from beginning to end. Rats are given one session daily of CDWM until they perform the task at a criterion level of at least 80% correct choices on two consecutive sessions.
  6. CDWM task training (during either single or dual-task training) includes an additional phase. After rats reach performance criterion on CDWM with short left and right goal arm inserts and the half sized central arm insert, the short goal arm inserts are removed and rats are trained without them, leaving only the first half of the stem cued via the central arm insert. The same performance criterion is used for this phase of training.
  7. Next, rats begin the dual-task phase, learning either the CDWM or CDSTANDARD task while also continuing to perform the other task (inter-task interval 15-20 min). In CDSTANDARD, floor inserts were identical to the CDWM floor inserts except for the size of the inserts. CDSTANDARD inserts span the length of the central stem and goal arms of the T-maze. The tasks are identical in every way except for the length of cues available to the animals. Rats are trained on one session of CDSTANDARD and one session of CDWM per day. Task type order during dual-task training is counterbalanced. We have found that rats are able to perform daily training sessions of 18 CDSTANDARD trials (9 wood trials and 9 mesh trials) and 18 CDWM trials (9 wood trials and 9 mesh trials) for a total of 36 daily trials. A typical dual task session takes ~1.5 h to run per rat. Performance criterion is again set at a performance level of at least 80% correct choices on two consecutive dual-task sessions.

Data analysis

To assess SWM ability, we compare choice accuracy of a single session and across all sessions between the two variants of the task. A selective performance accuracy deficit on CDWM indicates an SWM impairment, see Figure 7 below for performance data from one example dual-task session. We have used both between-subjects and within-subject designs.

Figure 7. Example of choice accuracy for one representative rat in a dual-task session. This rat performed 18/18 correct trials on CDSTANDARD and 12/18 correct trials on CDWM.


  1. All procedures are approved by the University of Delaware Institutional Animal Care and Use Committee.
  2. For an ongoing experiment, we trained 10 rats on both task variants, with half of the rats first learning CDSTANDARD, and then adding CDWM and the other half first learning CDWM alone, then adding CDSTANDARD. It took 18.8 (ranging from 11 to 29) sessions to reach criterion on CDSTANDARD and 51.5 (ranging from 35 to 78) sessions to reach criterion on CDWM. Rats that were first trained on CDSTANDARD then added CDWM took 34.2 sessions (range: 12 to 69 sessions) on the dual-task phase. Rats that were first trained on CDWM then added CDSTANDARD took an average of 13.8 (range: 11 to 18) sessions to reach criterion. In sum, CDWM takes 3-4 times longer to learn than CDSTANDARD. Additionally, training rats on CDWM in the single-task phase and adding CDSTANDARD in the dual-task phase gives us less variability in learning rates than training the rats on the tasks in the opposite order.


The authors would like to thank Dr. Mark Stanton for consulting on the development of the CD task and Gregory Peters, Crystal Shaw, and Henry Hallock, for working out the details of the task. This work was supported by the National Institutes of Health (R01 MH102394 to AG) and the Delaware Center for Neuroscience Research (P20 GM103653).


  1. Baddeley, A. (1992). Working memory. Science 255: 556-559.
  2. Fellows, B. J. (1967). Chance stimulus sequences for discrimination tasks. Psychol Bull 67: 87-92.
  3. Griffin, A. L., Owens, C. B., Peters, G. J., Adelman, P. C. and Cline, K. M. (2012). Spatial representations in dorsal hippocampal neurons during a tactile-visual conditional discrimination task. Hippocampus 22(2): 299-308.
  4. Hallock, H. L., Arreola, A. C., Shaw, C. L. and Griffin, A. L. (2013a). Dissociable roles of the dorsal striatum and dorsal hippocampus in conditional discrimination and spatial alternation T-maze tasks. Neurobiol Learn Mem 100: 108-116.
  5. Hallock, H. L. and Griffin, A. L. (2013). Dynamic coding of dorsal hippocampal neurons between tasks that differ in structure and memory demand. Hippocampus 23(2): 169-186.
  6. Hallock, H. L., Wang, A. and Griffin, A. L. (2016). Ventral midline thalamus is critical for hippocampal-prefrontal synchrony and spatial working memory. J Neurosci 36(32): 8372-89.
  7. Hallock, H. L., Wang, A., Shaw, C. L. and Griffin, A. L. (2013b). Transient inactivation of the thalamic nucleus reuniens and rhomboid nucleus produces deficits of a working-memory dependent tactile-visual conditional discrimination task. Behav Neurosci 127(6): 860-866.
  8. Shaw, C. L., Watson, G. D., Hallock, H. L., Cline, K. M. and Griffin, A. L. (2013). The role of the medial prefrontal cortex in the acquisition, retention, and reversal of a tactile visuospatial conditional discrimination task. Behav Brain Res 236(1): 94-101.
  9. Urban, K. R., Layfield, D. M. and Griffin, A. L. (2014). Transient inactivation of the medial prefrontal cortex impairs performance on a working memory-dependent conditional discrimination task. Behav Neurosci 128(6): 639-643. 


该协议描述了触觉视觉条件辨别(CD)T迷宫任务的两个变体的新颖的双重任务比较,一个取决于空间工作记忆(SWM; CDWM),另一个依赖于CDSTANDARD(CDSTANDARD)。 任务变体在其感官和运动要求以及大鼠的公开行为方面是相同的。 因此,依赖变量中的两个任务变体之间的差异,如选择准确性,神经激发模式,以及感兴趣脑区域的药理学或光遗传失活的影响可归因于SWM,排除了混淆感觉运动变量,如触觉, 视觉和自我运动线索。 CDWM任务协议在Hallock等人,2013b和Urban等人,2014年出版。
【背景】我们的实验室有兴趣探索工作记忆的神经机制。因此,我们开发了一个可用于评估大鼠空间工作记忆(SWM)能力的任务。工作记忆被定义为持有有限数量的“在线”信息,以便可以使用或操纵信息来指导目标导向的行为(Baddley,1992)。因为啮齿动物自然倾向于食用食物,它们是用于探测SWM的优秀模型。我们的实验室开发并使用了条件歧视(CD)T迷宫任务,其中纹理和颜色不同的地板镶嵌作为奖励目标臂的条件线索(Griffin等人,2012; Hallock和Griffin,2013; Hallock等等,2013a; Shaw等,2013; Hallock等,2016)。例如,如果老鼠遇到网眼插入物和右侧目标手臂遇到木质插入物时,学习选择左侧目标手臂。为了区分插入物,大鼠可以使用视觉信息(黑色与浅棕色),触觉信息(粗糙网格对平滑木材)或两种类型信息的组合。因为插入物覆盖迷宫的整个楼层,并且当大鼠进行目标臂选择时可用,所以该任务不需要SWM。最近,我们开发了一个工作记忆变体的工作(CDWM; Hallock等,2013b; Urban等,2014)。在这个任务的变型中,地板插入线索仅在迷宫的中央臂的一半上方延伸,并且当大鼠进行目标臂选择时不可用,因此需要老鼠保持提示的短暂时间时间为了做出正确的选择并获得食物奖励。在正在进行的实验中,我们发现可以在任务的两个变体上训练老鼠,为我们提供了一种强有力的方式来识别SWM的行为相关性,同时排除混淆感觉运动变量,如视觉,触觉和自我运动线索。

关键字:空间工作记忆, 条件辨别, T形迷宫, 编码, 回忆


  1. 男性长埃文斯连帽(哈兰,印第安纳波利斯)大鼠,在抵达时重量为250g至500g(约90天龄)
  2. 巧克力洒水用于食物奖励。我们的实验室使用厨师的品质品牌
  3. 70%乙醇用于清洁日常训练之间的迷宫


  1. 木制迷宫由中央杆(117 x 10 x 5厘米),两个目标臂(56.5 x 10 x 5厘米)和两个回弹臂(112 x 10 x 5厘米)组成。迷宫的地板用黑色乙烯基(3毫米厚)(图1)覆盖。


  2. 作为启动箱,附着在座椅上的具有塑料碟(直径:38cm)的木凳(高度:69cm)位于迷宫的底部。起始箱通过一个可拆卸的6厘米高的木质障碍物与迷宫分开
  3. 三个可移动的木地板插件,一面覆盖着黑色塑料网,另一侧则是平滑木材,作为条件线索(图2)。黑色网格用超滤胶粘在插入物的一侧,由4×4mm的网格正方形组成。对于任务的CDWM变体,中心臂插入物(74 x 8厘米)从启动盒的中间延伸到T形交叉点。对于任务的CD STANDARD 变体,中心臂插入物(117 x 8厘米)覆盖从起始框到T形交叉点的中心臂的整个长度。三个目标臂插入物,一个大插入物(61 x 8厘米)和两个小插入物(26 x 8厘米)放置在奖杯杯旁边的目标臂的末端

    图2.可拆卸的木制插入物,用于CDWM(A)和CD STANDARD (B)任务和关闭插入(C)
  4. 黑色的窗帘,围绕整个行为房间,距离迷宫51厘米远
  5. 远端线索录入黑色幕帘,离地板142厘米,位于起始台座后面,左右奖杯(图3)。提示是由彩色胶带制成的粉色三角形(38.1×29.2厘米),红色X(40.6×35.6厘米)和蓝色立方体(43.2×41.9厘米),


  6. 位于每个目标手臂末端的塑料杯(3厘米直径; 1厘米深),巧克力洒上(4-5件)。上限为20盎司。水/苏打瓶盖
  7. 一个60瓦的白炽灯附着在靠近天花板的后中间的窗帘轨道上。灯泡朝天花板朝上,因此房间昏暗,无法直接照亮迷宫的任何部分



图4.实验过程的流程图,从通过双重任务培训将老鼠驯化到生物群开始 >
  1. 将大鼠每笼笼养在温度和湿度控制的动物生物体中,并保持在受控的光周期(早上7:00至晚上7:00)。

  2. 在生殖器中经过7天的适应期后,将大鼠单独饲养并每天送至实验室5-7天,并由实验者每天处理15分钟。实验者通过在其膝盖上放置尿布并将动物放置在尿布的顶部来处理大鼠,允许动物自由移动。处理完毕后,实验者将一杯巧克力洒在家笼内。给大鼠20分钟吃掉,20分钟后取出杯子。然后将大鼠放入食物限制(每天4-5个食物颗粒),以将其保持在自由喂养体重的80%至90%之间,并随意获得水。大鼠从开始训练开始(,即处理时间)开始食用,直到测试结束。在处理的第一天记录起始体重,并在实验期间每周称量大鼠。
  3. 接下来,老鼠进行门柱训练。在这个阶段,将大鼠带到T迷宫所在的行为室。动物被限制在迷宫上的目标胳膊上,并允许从奖杯中吃巧克力,共进行6次试验(每个目标胳膊3次)。单次试验的持续时间为90秒,ITI为10秒。如果动物在不到90秒的时间内消耗了食物的报酬,试验将被提前终止。一旦老鼠连续两次在每次试验中消耗不到90秒的报酬,他们就会进入强制训练。动物通常需要3到5天才能达到目标箱训练的绩效标准。
  4. 强制运行会话由每天12次试验组成,其中6次左侧和6次正确试验以伪随机序列给出。在每次试验之前,实验者在其中一个目标臂的入口处放置一个可移动的木制障碍物。在试验开始时,大鼠被限制在起动箱与第二个木制可拆卸的屏障。强制运行的试验开始时,启动框障碍被删除。鼓励大鼠从启动箱,中臂下降,降低可用的目标手臂,消耗奖励,并通过返回臂返回起动箱。老鼠不要在迷宫的任何一点转过来,用木棍阻挡他的路。该路径修正程序仅在早期训练阶段才需要,之后实验者尽可能地尽可能地与大鼠进行交互。一旦老鼠连续两次在80%的强制运行试验中获得奖励,他们就进入了单一任务阶段的培训。强制训练通常持续6至7天。
  5. 大鼠训练CDWM任务(图5)或CD STANDARD 任务(图6)。起始任务变体在大鼠之间平衡,得到两组; CDWM组和CD STANDARD 组。

    图5. CDWM任务示意图。 可拆卸的木质插入物,一侧有塑料黑色网眼,另一侧具有光滑的木材,线中央臂的前半部分。见视频1.

    图6. CD STANDARD 任务的示意图。 可拆卸的木制插件一侧用塑料黑色网眼覆盖,另一侧的光滑木材覆盖整个中央手臂和目标手臂。见视频2.

    1. 在每次试验之前,实验者将地板插入物放置在T型迷宫的中心杆和两个目标臂上。两个短的左,右目标臂插入物和半尺寸的中心杆插入物将具有相同的颜色/纹理。目标插入物用于帮助大鼠学习和维持条件歧视提示与食物奖励的位置(例如Wood-Right或Mesh-Left)的关联。我们的推理是,缺乏或存在食物会增加有条件歧视的学习,如果伴随着条件提示的呈现。在CDWM任务中,插入是CD STANDARD 插入的大小的一半。中央臂插入物覆盖迷宫中央杆的前半部分。目标臂插入覆盖距离T形交叉点最远的目标臂的最后一半。迷宫插入物的一面被黑色网状物覆盖,另一面是光滑的木材(浅棕色)。我们发现没有必要将可拆卸的插件固定到迷宫。这些插件紧贴在迷宫中,所以每次试验都将它们放在相同的位置。
    2. 在这两个任务中,条件提示与T-maze上的食物奖励的位置相关联需要根据地板插入物的纹理和颜色(平滑木或黑色网格)选择左侧或右侧目标。例如,当他们经历光滑的木材和左转时,老鼠学会在迷宫的T形交叉处右转,当他们经历黑色网格时。试验以每场相同数目的网格和木材试验的伪随机序列呈现(Fellows,1967)。奖励意外事件在大鼠间平衡,实验中有一半的老鼠学习在网格试验中选择正确的目标胳膊,而在左侧目标手臂上进行木材试验,另外一半的老鼠学习相反的规则网格,就在木头上)。即使在连续试验中提供相同的提示,每次试验之间都会翻转提示插入。翻转提示插入是为了确保大鼠不能通过使用听觉线索来解决任务。
    3. 在每次试验之间,老鼠在起动箱中等待20秒,而实验者准备下一次试验的迷宫。由24个试验组成的典型的训练课程(12个木材12目)从头到尾大约需要30-40分钟。每周给予大鼠CDWM一次会议,直到他们在两个连续会话上至少80%正确选择的标准级别执行任务。
  6. CDWM任务培训(在单任务或双任务培训期间)包括一个附加阶段。老鼠达到CDWM的性能标准后,左侧和右侧目标臂插入物和半尺寸的中心臂插入物,短针臂插入物被移除,大鼠训练没有它们,仅通过中央臂仅留下茎的前半部分插。这个训练阶段使用相同的绩效标准。
  7. 接下来,老鼠开始双重任务阶段,学习CDWM或CD STANDARD 任务,同时继续执行其他任务(任务间隔15-20分钟)。在CD STANDARD 中,地板插入物与CDWM地板插入件相同,除了插入件的尺寸。 CD 标准插入跨越T迷宫的中心杆和目标臂的长度。除了动物可用的线索的长度之外,任务都是相同的。大鼠在一天的CD标准和每天一次的CDWM课程中进行培训。双重任务培训期间的任务类型顺序是平衡的。我们发现大鼠能够进行18次CD标准试验(9次木试验和9次试验)和18次CDWM试验(9次木试验和9次试验)的日常训练,总共36次日常试验。典型的双重任务每个大鼠需要1.5小时运行。性能标准再次设置在两个连续双重任务会话上至少80%正确选择的性能级别。
  8. 数据分析

    为了评估SWM能力,我们比较了单个会话的选择准确性和任务的两个变体之间的所有会话。 CDWM上的选择性绩效准确度赤字表示SWM损害,参见下面的图7从一个示例性双任务会话的性能数据。我们已经使用主题和主题内设计。

    图7.双任务会话中一个代表性大鼠的选择准确性示例。 该大鼠对CDWM的CD 标准和12/18正确试验进行了18/18次正确试验。


    1. 所有程序均由特拉华州大学体育动物护理和使用委员会批准。
    2. 对于正在进行的实验,我们在两个任务变体上训练了10只大鼠,其中一半的大鼠首先学习CD STANDARD ,然后单独添加CDWM和另一半首先学习CDWM,然后添加CD < STANDARD 。共有18.8次(从11到29次),达到CD标准的标准,51.5(范围从35到78),达到CDWM的标准。第一次接受CD STANDARD 训练的大鼠,则在双重任务阶段增加了CDWM,占用了34.2个会话(范围:12至69次)。首先在CDWM上训练的大鼠然后添加CD 标准平均达到13.8(范围:11至18),达到标准。总之,CDWM比CD STANDARD 需要3-4倍的时间才能学习。此外,在双重任务阶段,在单任务阶段对CDWM进行培养大鼠,并在双重任务阶段添加CD STANDARD ,使我们在学习率上的变化不大,而不是按照相反的顺序对老鼠进行任务训练。 />


    作者感谢Mark Stanton博士就光碟任务的发展进行磋商,并就格雷戈里·彼得斯(Crystal Shaw)和亨利·哈洛克(Henry Hallock)作了详细的工作。这项工作得到了国立卫生研究院(R01 MH102394,AG)和特拉华神经科学研究中心(P20 GM103653)的支持。


    1. Baddeley,A。(1992)。  工作记忆。 科学 255:556-559。
    2. 研究员BJ(1967)。机会刺激歧视任务的序列。 Psychol Bull 67:87-92。
    3. Griffin,AL,Owens,CB,Peters,GJ,Adelman,PC and Cline,KM(2012)。< a class ="ke-insertfile"href ="http://www.ncbi.nlm.nih.gov/pubmed/21080411"target ="_ blank">在触觉视觉条件歧视任务期间背侧海马神经元中的空间表示。海马 22(2):299-308。 >
    4. Hallock,HL,Arreola,AC,Shaw,CL和Griffin,AL(2013a)。< a class ="ke-insertfile"href ="http://www.ncbi.nlm.nih.gov/pubmed/23261856 "target ="_ blank">背部纹状体和背侧海马在条件歧视和空间交替T迷宫任务中的分离角色。 Neurobiol Learn Mem 100:108-116。 >
    5. Hallock,HL和Griffin,AL(2013)。动态在结构和记忆需求不同的任务之间编码背部海马神经元。海马 23(2):169-186。
    6. Hallock,HL,Wang,A.和Griffin,AL(2016)。腹侧中线丘脑对于海马前额叶同步和空间工作记忆至关重要。 J Neurosci 36(32):8372-89。
    7. Hallock,HL,Wang,A.,Shaw,CL和Griffin,AL(2013b)。< a class ="ke-insertfile"href ="http://www.ncbi.nlm.nih.gov/pubmed/24341710"target ="_ blank">丘脑核复原和菱形核的瞬时灭活会导致工作记忆依赖的触觉条件鉴别任务的缺陷。 Behav Neurosci 127(6) :860-866。
    8. Shaw,CL,Watson,GD,Hallock,HL,Cline,KM and Griffin,AL(2013)。  内侧前额叶皮质在获取,保留和逆转触觉视觉空间条件歧视任务中的作用。 Behav Brain Res 236( 1):94-101。
    9. Urban,KR,Layfield,DM and Griffin,AL(2014)。 
  • English
  • 中文翻译
免责声明 × 为了向广大用户提供经翻译的内容,www.bio-protocol.org 采用人工翻译与计算机翻译结合的技术翻译了本文章。基于计算机的翻译质量再高,也不及 100% 的人工翻译的质量。为此,我们始终建议用户参考原始英文版本。 Bio-protocol., LLC对翻译版本的准确性不承担任何责任。
Copyright: © 2017 The Authors; exclusive licensee Bio-protocol LLC.
引用: Readers should cite both the Bio-protocol article and the original research article where this protocol was used:
  1. Edsall, A., Gemzik, Z. and Griffin, A. (2017). A Tactile-visual Conditional Discrimination Task for Testing Spatial Working Memory in Rats. Bio-protocol 7(10): e2282. DOI: 10.21769/BioProtoc.2282.
  2. Hallock, H. L., Wang, A. and Griffin, A. L. (2016). Ventral midline thalamus is critical for hippocampal-prefrontal synchrony and spatial working memory. J Neurosci 36(32): 8372-89.