Summary

Fear Incubation Using an Extended Fear-Conditioning Protocol for Rats

Published: August 22, 2020
doi:

Summary

We describe an extended fear-conditioning protocol that produces overtraining and fear incubation in rats. This protocol entails a single training session with 25 tone-shock pairings (i.e., overtraining) and a comparison of conditioned freezing responses during context and cue tests 48 h (short-term) and 6 weeks (long-term) after training.

Abstract

Emotional memory has been primarily studied with fear-conditioning paradigms. Fear conditioning is a form of learning through which individuals learn the relationships between aversive events and otherwise neutral stimuli. The most-widely utilized procedures for studying emotional memories entail fear conditioning in rats. In these tasks, the unconditioned stimulus (US) is a footshock presented once or several times across single or several sessions, and the conditioned response (CR) is freezing. In a version of these procedures, called cued fear conditioning, a tone (conditioned stimulus, CS) is paired with footshocks (US) during the training phase. During the first test, animals are exposed to the same context in which training took place, and freezing responses are tested in the absence of footshocks and tones (i.e., a context test). During the second test, freezing is measured when the context is changed (e.g., by manipulating the smell and walls of the experimental chamber) and the tone is presented in the absence of footshocks (i.e., a cue test). Most cued fear conditioning procedures entail few tone-shock pairings (e.g., 1-3 trials in a single session). There is a growing interest in less common versions involving an extensive number of pairings (i.e., overtraining) related to the long-lasting effect called fear incubation (i.e., fear responses increase over time without further exposure to aversive events or conditioned stimuli). Extended fear-conditioning tasks have been key to the understanding of fear incubation’s behavioral and neurobiological aspects, including its relationship with other psychological phenomena (e.g., post-traumatic stress disorder). Here, we describe an extended fear-conditioning protocol that produces overtraining and fear incubation in rats. This protocol entails a single training session with 25 tone-shock pairings (i.e., overtraining) and a comparison of conditioned freezing responses during context and cue tests 48 h (short-term) and 6 weeks (long-term) after training.

Introduction

Memory is a psychological process encompassing different phases: information acquisition, consolidation (allows for the stability of acquired information), and retrieval (evidence for the consolidation process)1. During the consolidation phase, the establishment of new synaptic connections and modification of pre-existing connections occur. This suggests the necessity for a period of time during which molecular and physiological events responsible for these changes occur1,2. These physiological or molecular changes vary whether the retrieved events are emotionally charged or not (i.e., emotional memory). For instance, research has shown that the lateral nucleus and basolateral amygdala complex are particularly relevant to emotional memory3,4,5.

Emotional memory phenomena have been primarily studied with fear conditioning paradigms5,6. Fear conditioning is a form of learning through which individuals learn the relationships between aversive events and otherwise neutral stimuli7. Fear conditioning paradigms produce molecular, cellular, and structural changes in the amygdala. In addition, fear conditioning modifies the connectivity of the hippocampus during the consolidation and retrieval processes of emotional memory.

One of the most commonly used procedures for studying fear memories is classical (Pavlovian) conditioning in rats. This procedure typically uses footshock (US) as the aversive stimulus, which is delivered once or several times across one or several sessions. The conditioned response (CR) of rats exposed to this procedure is freezing (i.e., “generalized immobility caused by a generalized tonic response of the animals’ skeletal musculature except those muscles used in breathing”7 ). This response could be assessed on two types of tests: context and cue tests. For the context test, the subject undergoes a given number of footshocks during the training session, and then is removed from the experimental chamber for a defined time. During the test, the subject is returned to the same context in which the training took place and different measures of freezing are collected in the absence of footshocks (e.g., duration, percentage or frequency of freezing episodes), and compared to baseline levels established during the training phase. For the second type of test, cue test, a stimulus (typically a tone) is paired with the footshocks during the training phase (i.e., conditional stimulus, CS). After training is completed, the animal is removed from the training context for a defined time and is subsequently placed in a modified context (e.g., a different experimental chamber that has different shapes of walls and different smell). The cue is then presented a given number of times, and freezing responses to the cue are measured and compared to baseline levels collected during training. The most common version of this paradigm uses 1 to 3 tone-shock pairings during a single training session, followed by context and cue tests conducted a number of hours or a few days later.

Other less frequently implemented fear conditioning procedures involve an extensive number of shock-cue pairings (i.e., trials), which have often been called overtraining procedures8. A growing interest in these tasks is related to their long-lasting and increased memory effects called fear incubation (i.e., conditioned fear responses increase over time in the absence of further exposure to aversive events or conditioned stimuli)9,10,11. An example of such overtraining procedures entails a training phase of 100 tone-shock pairings distributed across 10 sessions, followed by context and cue tests conducted 48 h and 30 days later11,12. To avoid extensive training spread across several days, Maren (1998) reported that overtraining could be established and optimized in a single session with 25 pairings8. The incubation effect is evidenced in significantly higher levels of conditioned fear in rats tested 31 days after training, as compared to rats tested 48 h after. Extended fear-conditioning tasks have been key for the understanding of behavioral and neurobiological aspects underlying fear incubation, including its relationship with other psychological phenomena (e.g., delayed-onset post-traumatic stress disorder)11,12,13.

Here, we describe an extended fear-conditioning protocol that induces overtraining and fear incubation in rats. Different to other paradigms that require several days of training11, the current protocol is focused on a single training session8. We used 25 tone-shock pairings to produce higher conditioned freezing responses during context and cue tests conducted 6 weeks after training, as compared to tests conducted 48 h after.

Protocol

The following protocol was approved by the Institutional Animal Care and Use Committee of Fundación Universitaria Konrad Lorenz (IACUC-KL). The universal declaration of animal rights issued by International League of Animal Rights, Geneva, Switzerland (1989), and ethical principles of experimentation with animals issued by ICLAS were respected.

1. Subject preparation

  1. Select male adult Wistar rats (n = 12). House them in groups of four per cage for three days of acclimatization, prior to the beginning of the training and testing protocol. Provide rats with free access to water throughout the experiment. Control the room temperature between 20 °C to 25 °C, under a 12 h light-dark cycle (lights on at 07:00 h).
    NOTE: Rat strains had shown differential performance during fear conditioning. For instance, Schaap et al. (2013) reported that Wistar and Lewis strains showed longer durations of freezing behavior compared with Fawn Hooded and Brown Norway rats12. Thus, differences in pain and thermal threshold should be assessed to adjust the intensity and duration of shocks.
  2. Maintain rats at 85% of their free-feeding weights (normal weight between 350-400 g) by giving restricted food access at the same hour every day. Weigh rats every day at the same hour during the light cycle. Calculate the ad lib weight (100% weight) for three days before the start of extended fear-conditioning training.
    NOTE: Animals used in the present experiment were tested on additional instrumental tests that are not reported in here. Food deprivation was required for those additional tests. This procedural variation is assumed as likely to expand the scope of the present procedure, as it suggests the potential for instrumental-fear combined tests. However, studies using only fear conditioning tests will not require food deprivation.
  3. Randomly assign subjects to one of the following groups: emotional testing 6 weeks after training (n = 6); emotional testing 48 h after training (n = 6).
  4. Perform training and tests at similar hours, during the light phase of dark-light cycle. Assign the animals to the same experimental chamber and maintain the same order of animals during training and testing.
    NOTE: An additional control that could be implemented is counterbalancing the order of animals during training and testing phases. We recommend using this technique when multiples groups are assessed, or different tasks are applied across experiments, to reduce a possible effect of task-order on behavior.

2. Apparatus setting and shock calibration

  1. Clean all the internal surfaces of the experimental chamber and stainless-steel grid floor with 10% ethanol. Repeat before testing each animal.
  2. Connect the equipment to a computer using a USB cable and start the freezing detection system equipment: the CPU, the controlling cabinet, the infrared light, the aversive stimulator/scrambler, and the shock-intensity calibrator.
    NOTE: Although this protocol was executed using commercially available instruments (Table of Materials), equipment and software of different brands can be used. The apparatus consists of an internal acrylic square chamber (29.53 cm x 23.5 cm x 20.96 cm, called the experimental chamber) embedded in a wooden box covered with plastic formic. The external doors allow the isolation of sound, noise or light (attenuating box doors). The camera is located laterally in the internal part of the external door. The internal acrylic box with floor metal grids (36 stainless-steel rods, each one of 3 mm diameter and spaced 8 mm, center to center) allows footshock delivery. In one of the internal-lateral walls, a speaker is located 6 cm from the floor to present an auditory cue.
  3. Connect the red and black clips of the shock intensity calibrator (i.e., positive and negative connectors) to two any different rods on the grid floor. Connect the USB cable to the corresponding port of the computer. Make sure to connect the red and black clips to bars separated by another bar.
    NOTE: This section describes the shock intensity calibration process using a specific brand of equipment mentioned in the Table of Materials. However, the calibration process can be performed using different brands of equipment. It is recommended to calibrate the intensity of the shock in three sectors of the grid floor to verify that it is consistent. In addition, always remove fecal and urine residues from the grid floor to avoid interference during the delivery of the shock.
  4. Start the shock-intensity calibrator software (Table of Materials). Choose an intensity of 1.0 mA in the application by clicking on the range arrow. Then, change the Run/Stop switch to Run.
    NOTE: We propose 1.0 mA based on our studies with rodent models in our lab and literature that reports a range from 0.75 mA to 1.5 mA as adequate for studies of fear conditioning33,34,35.
  5. Switch on the aversive stimulator or the equipment used to deliver the footshocks and look at the shock intensity displayed on the panel of the application. If needed, adjust the intensity to 1.0 mA using the knob on the aversive stimulator.
    NOTE: Aversive stimulator should be set to “OUT” to appropriately test, calibrate, and run the experiment.

3. Freezing detection system calibration

  1. Close the experimental chamber and sound-attenuating box doors. Do not introduce the animal at this point, as it will be placed into the chamber after the freezing detection system calibration has been completed. Check that the light intensity inside the box is between 20 and 30 lux.
  2. Start the freezing detection system software and open the Experiment setup dialogue window. Enter the details of each subject (such as subject identification number, date and group) and load the file titled “Training protocol VFC.pro” (available at http://doi.org/10.17605/OSF.IO/4NKFQ).
    NOTE: Context and cue tests use a different program configuration; thus, make sure to use the correct file on each test. At this point the correct file corresponds to “Training protocol VFC.pro”. Remember that during test phases the file corresponding will be different to training session.
  3. Choose the corresponding camera(s) and check the Save Video option (if needed). Set the Motion Threshold to 100, and Min Freeze Duration to 30 frames.
    NOTE: This Motion Threshold value is based on the size of the species used (based on number of pixels). Minimum Freeze Duration value is recommended by the manufacturer. These values are used to ensure proper recognition of the animal in the chamber.
  4. Verify that the live feed from the chosen camera(s) appears on the screen, together with the motion threshold graph and the timeline of the different stimuli that are presented during the training (e.g., sound and shock).
    NOTE: Using a different brand, the equipment setup should offer the possibility to measure the movements of the animal to detect an “index” of motion that should allow comparisons on the amount of time the animal is moving or freezing. Another possibility is using a software that with only the video source (during or after the experiment) can detect the amount of time the animal is in motion or freezing, such as free software ImageFZ13, open-source toolbox in Matlab14, or a free classifier of animal behavior as JAABA15.
  5. Click the Calibrate option three times, while checking that the Motion Index remains below 100 (threshold). Then, set the equipment to lock by clicking on the corresponding button on the screen.
    NOTE: This section describes a freezing detection system calibration process using a specific brand of equipment listed in the Table of Materials. As was mentioned before, the calibration process can be conducted using different brands of equipment (for a review of different options in equipment and software see Anagnostaras et al. 2010)16.

4. Extended fear conditioning training

  1. Transport the rats in their home cages, covered with a cloth, from the animal care facility to the behavioral training room in the laboratory. Avoid exposure to noise or stress-generating conditions during the transport of animals to the behavioral training room. If several animals are transported at the same time, only bring the animals to be tested and maintain other rats in a holding room to enhance experimental control. Gently handle the animals for 2 min before starting the training.
    NOTE: In the protocol, the animals were handled each day for 2 minutes before behavioral training. Following handling, animals were introduced in the experimental chamber. We recommended to manipulate animals to make rats habituate to the researcher.
  2. Introduce the rat to the experimental chamber. Handle it gently by the base of its tail and place it on the middle of the chamber. Close the experimental chamber and sound-attenuating box doors.
  3. Start the session by clicking on the Record button. Let the rat acclimate to the chamber for 3 min. This 3 min period is the standard recommended by the equipment manufacturer and serves as a baseline and habituation time to the chamber.
  4. Deliver twenty-five tone-shock pairings (trials) with a 60 s Inter-Trial Interval (ITI), starting on minute 3 of the session. Present the tone (conditioned stimulus – CS; 90 dB SPL, 2000 Hz, 50-ms Rise Time) during the last 10 s of each ITI, and the shock (unconditioned stimulus – US) during the last 2 s of each ITI.
    NOTE: Activation of the Record button is conditional on cameras being properly calibrated and locked.
  5. Remove the rat from the experimental chamber when the 28 min session is over. Return animals to the respective home cage. Transport the rats in their home cages covered with a cloth from the behavioral training room to the animal care facility.
  6. Repeat freezing detection system calibration (steps 3.1-3.5) and fear conditioning (steps 4.1 and 4.3) to train all the subjects.
    NOTE: We strongly recommend recalibrating the detection system for each animal to ensure that the software maintains the same parameters when it processes information for freezing detection.
  7. Resting period: During this period, have the animals rest individually in their home cages. Monitor the weight of the animals twice per week during the 6 weeks of the incubation period. Gently manipulate each animal for two min while they are weighted.

5. Context test – single 10 min session

  1. After the training phase, expose the animals to the first memory test called context test. During this 10 min phase, expose the rats to the same context in which training took place but do not present cues or shocks. Transport the rats in their covered home cages (e.g., with a cloth) from the animal care facility to the behavioral training room. Keep in mind that animals were divided into groups, thus one group is tested 48 h after the training phase and the other group is tested 6 weeks after training (see Figure 1).

Figure 1
Figure 1: Timeline of the experiment. Please click here to view a larger version of this figure.

  1. Clean all the internal surfaces of the experimental chamber and stainless-steel grid floor with 10% ethanol. Repeat before testing each animal.
  2. Repeat freezing detection system calibration (steps 3.1 to 3.5). Open the Experiment setup dialogue window and load the file named “Context test protocol.pro”, which is available from http://doi.org/10.17605/OSF.IO/4NKFQ.
    NOTE: This file contains the setup for this experimental phase that consists of no shocks or tones.
  3. Introduce the animal to the experimental chamber. Handle it gently by the base of its tail and place it on the middle of the chamber. Close the experimental chamber and sound-attenuating box doors.
  4. Start the session by clicking on the Record button. During this single 10 min context-test session, no stimuli are presented (shock neither sound).
  5. Remove the subject from the experimental chamber when the 10 min session is over. Return the animals to their respective cages and transport the rats in their covered home cages from the behavioral training room to the animal care facility. Repeat steps 5.2-5.5 to test all the subjects.

6. Cue test – single 13 min session

  1. One day after the context test, have animals undergo the second test of memory called cue test. During this phase, the rats will be in a different context of training during 13 min.; cues (tones) are presented, but no shocks are delivered. Transport the rats in their home cages covered with a cover from the animal care facility to the behavioral training room. Test a group 72 h after the fear conditioning training, and test another group 6 weeks and one day after training (Figure 1).
    NOTE: A different system of transportation (from the animal care facility to the experimental room) could be implemented to differentiate more the context and cue tests. Since the animals were transported to the training session and context test session in their home cages, a different transport cage, bedding and/or cover could be used during transportation to the cue test session.
  2. Clean all the internal surfaces of the experimental chamber and stainless-steel grid floor with 10% ethanol. Repeat before testing each animal.
  3. To change the visual context, insert the plastic surrounding wall to the experimental chamber.
  4. To change olfactory context, apply 1% acetic acid to a cotton-tipped swab, and place it in the metal tray below the grid floor17,18,19.
  5. Repeat the freezing detection system calibration (steps 3.1-3.5). Load the file named file “Cue test protocol.pro” file, which is available from http://doi.org/10.17605/OSF.IO/4NKFQ.
    NOTE: This file contains the setup for this experimental phase, which consists of delivery of the same tones presented during the training phase (CS), but in the absence of shocks (US).
  6. Introduce the animal to the experimental chamber. Handle it gently by the base of its tail and place it on the middle of the chamber. Close the experimental chamber and sound-attenuating box doors.
  7. Start the session by clicking on the Record button. During the single 13 min cue test session, the CS stimulus (tone) is presented 10 times, starting on minute 3 of the session.
    NOTE: The first 3 min correspond to the baseline of this session, followed by 10 cue test trials (that is, 10 s each) delivered with 50 s ITIs in the absence of shocks. The delivery of tones is automatic, via using the previously loaded file.
  8. Remove the animal from the experimental chamber when the 13 min session is over. Return animals to the respective cage and transport them covered to the animal care facility. Repeat steps 6.2 through 6.5 to test all the subjects.

7. Data analysis

  1. Obtain the general activity index (i.e., motion index) that is derived from the video stream using the freezing detection system software. This software automatically transforms the motion index to provide the percentage of freezing time per session and the number of freezing episodes. Set the freezing threshold to the default Minimum Freeze Duration setting of the system (1 s = 30 frames).
  2. Use the additional custom-made program (file available from http://doi.org/10.17605/OSF.IO/4NKFQ) to obtain:
    1. Use the program to determine the percentage of freezing during the first three minutes of the training session (i.e., baseline freezing, since no shocks or tones were presented before or during that 3 min period) and during the first three minutes of the cue test session.
    2. Use the program to determine the percentage of freezing for each of eight 3 min bins of the training session.
    3. Use the program to determine the percentage of freezing during the cue presentations (i.e., freezing during the tones) and no-cue periods (intertrial intervals; ITIs), for both training and cue-test sessions.
  3. To obtain these data, open the freezing detection system software.
    1. Select File | Reports | Batch Component summary.
    2. Select the file with extension .CMP available from http://doi.org/10.17605/OSF.IO/4NKFQ.
    3. Name the output file and change Motion Threshold to a 100. Then, click OK.
    4. Select the files to be analyzed (extension .RAW). These files are automatically generated from the freezing detection system software when the session is over and correspond to the raw data of each session. Initially, the files are saved in the desktop of the computer, but they can be stored in a custom folder (e.g., Documents-Fear conditioning) to facilitate their subsequent identification and opening when they need to be analyzed.
    5. Open the output files (extension .CSV). They can be edited in a spreadsheet software for further analysis. This file contains the results of freezing during the experimental session.
      NOTE: To obtain the total percentage of freezing, divide the time that the subject spent immobile over the total session time. The number of freezing episodes can be calculated counting the number of freezing events through the session. In both cases, it is necessary to define a motion threshold based on a minimum freeze duration. This is the temporal criterion that defines whether a Freeze Episode is recorded. Automated systems of recording can use certain amount of frames per second (fps) as a measure of minimum freeze duration. For instance, with a sample rate of 30 fps, a minimum freeze duration of 15 frames will record as freezing an instance of immobility that last for 30 s.
  4. Calculate the average duration of each freezing episode for each session (training and both tests, context and cue) by dividing the total freezing duration (in seconds) over the total number of freezing episodes.

Representative Results

Variations in percentage of freezing time during different stages of the training session were analyzed for all subjects (n = 12) using a dependent t test (Table 1). Animals were active and explored the experimental chamber during the first three minutes of the training session (first day of the protocol), time during which no tones or shocks were delivered (i.e., baseline-BL). As shown in Figure 2A, percentage of freezing time during the subsequent 25 tone-shock pairings (M = 48.88; SE = 4.37) was significantly higher than during BL (M = 14.65; SE = 4.05), which is assumed as an indication of fear acquisition.

Statistic Test Figure Phases
Dependent t Test 2A t (11) = -6.21, p < .001, d = 2.34
3-min bins
Repeated Measures ANOVA 2B F (3.75, 41.32) = 11.19, p < .001, ηp2 = .50.
Phases Group Phases X Group
Mixed ANOVA 2C F(3, 30) = 14.21, p < .001, ηp2 =.58 F(3, 30) = 4.63, p < .05, ηp2 =.31 F(1, 10) = 2.06, p >.05, ηp2 =.17
One-Way ANOVA 3A F(1, 10) = 6.91, p < .05, ηp2 =.40
One-Way ANOVA 3B F(1, 10) = 10.30, p < .05, ηp2 =.50
One-Way ANOVA 3C F(1, 10) = 5.83, p <. 05, ηp2 =.36
Mixed ANOVA 4A F(2, 20) = 29.28, p < .001, ηp2 =.74 F(2, 20) = 2.33, p >.05, ηp2 =.18 F(1, 10) = 2.14, p >.05, ηp2 =.17
Mixed ANOVA 4B F(1, 10) = 1.53, p >.05, ηp2 =.13 F(1, 10) = 3.98, p < .05, ηp2 =.28 F(1, 10) = .23, p >.05, ηp2 =.02
Mixed ANOVA 4C F(1, 10) = 25.43, p < .001, ηp2 =.71 F(1, 10) = 6.17, p < .05, ηp2 =.38 F(1, 10) = .22, p >.05, ηp2 =.02

Table 1: Statistics used in the data analysis. For Figure 2A, mean percentage of freezing of all subjects (n = 12) during the first 3 min of the training session (corresponding to baseline, BL) was compared to the percentage of freezing during the remaining 25 min of the session (25 tone-shock trials) showing a significant difference and a large effect size (Cohen’s d = 2.34). For Figure 2B, a comparison was performed across bins of 3 minutes showing a significant difference in a Repeated Measures ANOVA test (BL and eight 3 min bins). For Figure 2C, comparisons between the mean percentage of freezing of each group of rats during the baseline (BL, first 3 min of the training session), training period (25 tone-shock pairings), context test session, and cue test session were conducted via a Mixed ANOVA with between-subjects factor the group (48 h or 6 weeks) and within-subjects factor the phases (BL, Training, Context Test and Cue Test). Differences in phases and group, but not in the interaction Phases*Group were found. Figure 3A – 3B shows data on activity (panel 3A, motion index), freezing (panel 3B, mean freezing in seconds) and duration of episodes (panel 3C, mean freezing episodes in seconds). These data were analyzed using a One-way ANOVA, which indicated differences between groups in all measurements. Finally, for Figure 4A – 4C a Mixed ANOVA was performed for each panel (A, B and C), having as between-subjects factor the group (48 h or 6 weeks) and within-subjects factor the phases (BL, Training, Context Test and Cue Test).

Figure 2
Figure 2: Training phase of an extended cued fear conditioning protocol. Data are shown as the mean (bars) and the SEM (error bars) of the freezing response. (A) shows mean percentage of freezing of all subjects (n = 12) during the first 3 min of the training session, during which no shocks or tones were presented (baseline, BL), and the remaining 25 min of the session (25 tone-shock trials, with intertrial interval, ITI, of 60 s); *** = different from BL (p < .001). (B) shows average freezing time of all the animals (n = 12) during the 3 min baseline period (BL, no shocks or tones delivered) and subsequent 3 min bins of the training session; *** = different from all the remaining bins (p < .001). (C) shows mean percentage of freezing of each group of rats (testing 48 h after training; testing 6 weeks after training) during the baseline (BL, first 3 min of the training session), training period (25 tone-shock pairings), context test session, and cue test session; * = different from testing after 48 h (mean diffContext = -34.95, SE = 14.99, p < .05, Cohen´s d = 1.34); a = different from training period (mean diffTraining48h = 42.51; SE = 7.28; p < .05; Cohen´s d = 3.03); b = different from training period (mean diffTraining6Weeks = 25.94; SE = 7.28; p < .05; Cohen´s d = 1.77), context test (mean diffContext6Weeks = 50.36; SE = 10.58; p < .01; Cohen´s d = 3.13), and cue test (mean diffCue6Weeks = 55.86; SE = 10.25; p < .01; Cohen´s d = 2.47). Please click here to view a larger version of this figure.

An analysis of the freezing response throughout acquisition was conducted by segmenting the training session in eight 3 min bins (Figure 2B). These data show that the mean time allocated to this response reaches asymptote near or at 180 s during the first three tone-shock trials (i.e., Bin 1). This finding has been considered in previous research an indication of overtraining11. Repeated-measures ANOVA revealed consistent significant differences between baseline and all subsequent bins, with large effect sizes (Table 1 and Table 2).

Comparison Mean difference Standard error p value Cohen´s d
Bin baseline vs bin 1 -60.075* 12,243 < .05 1.95
Bin baseline vs bin 2 -69.053* 16,220 < .05 1.89
Bin baseline vs bin 3 -66.197* 13,706 < .05 1.91
Bin baseline vs bin 4 -68.595* 11,969 < .05 2.08
Bin baseline vs bin 5 -65.475* 10,991 < .05 2.15
Bin baseline vs bin 6 -65.795* 13,509 < .05 2.06
Bin baseline vs bin 7 -72.900* 12,231 < .05 2.53
Bin baseline vs bin 8 -78.633* 8,692 < .001 3.37

Table 2: Mean difference, standard error and effect size for 3 min bins in Figure 2B. This table shows the comparisons between the baseline Bin and each of the subsequent bins (Figure 2B). Mean difference, standard error, and p-value and Cohen’s d are reported as an index of the size of these differences (effect size).

A mixed ANOVA was conducted to test differences in percentage of freezing during the task, having phases (BL, training, context test, and cue test) as the within-subject factor and group (48 h and 6 weeks) as the between-subjects factor (Table 1). Percentage of freezing of all animals during the training period was significantly higher than during the baseline period (see Figure 2C). No significant differences were observed between percentage of freezing during the memory tests and the training period (ps > .05).

No significant differences between the two groups (48 h and 6 weeks) were observed in the percentage of freezing during BL, training, and cue test (ps > .115; see Figure 2C). Conversely, animals tested 6 weeks after training showed significantly higher percentages of freezing during the context test than animals tested at 48 h, with a large effect size (see Figure 2C). Overall, Figure 2C shows that freezing during long-term delayed context and cue tests (i.e., 6 weeks after training) was overall significantly higher than during the training session. The opposite declining trend was observed in the group of animals that were tested 48 h after training. However, these differences in the group of 48 h were not statistically significant (ps > .05). Finally, although the freezing level showed differences across different phases, they could be considered low compared to other protocols. One explanation could be inherent methodological differences between laboratories or the motion index threshold established during the calibration process, making comparison of data among laboratories difficult.

The conditioned freezing response of the two groups of subjects during the context test was further explored via analysis of other measures, namely average activity (i.e., motion index), total freezing time and freezing time per episode. A one-way ANOVA was used to test differences across these variables (Table 1). Activity of subjects that were tested 6 weeks after training was significantly lower than that of animals tested 48 h after the training session (Figure 3A). Accordingly, total freezing time of animals tested shortly after training was significantly lower than that of animals tested 6 weeks after (Figure 3B). Lastly, an analysis of the average duration of each freezing episode indicated that animals tested 6 weeks after training displayed longer freezing episodes than animals tested 48 h after training (Figure 3C). Altogether, these findings indicate a fear incubation effect.

Figure 3
Figure 3: Effects of an extended cued fear conditioning protocol on freezing response of rats.
Data are shown as the mean (bars) and the SEM (error bars) of the freezing response. (A) shows activity (i.e., motion index) of each group of subjects (testing 48 h after training; testing 6 weeks after training) during the context test; * = different from 6 weeks. (B) shows the average total freezing time (in seconds) of each group of subjects during the context test; * = different from 6 weeks. (C) shows the average duration of each freezing episode (in seconds) for each group of subjects during the context test; * = only different from 6 weeks. Please click here to view a larger version of this figure.

A further examination of performance during the cue test session was conducted via analyses of (a) percentages of freezing during baseline periods (BL Training and BL Cue Test) and during the entire 10 min cue test (ten 10 s tone presentations and ten ITIs of 50 s – Figure 4A), (b) average freezing time specifically during the 10 s presentations of the cue (tone), for both Training and Cue Test sessions (Figure 4B), and (c) average freezing time (in seconds) during the 50 s intertrial intervals (ITIs; i.e., no-tone periods only – Figure 4C). A mixed ANOVA was used to analyze each of these dependent measures, assuming phases (BL Training, BL Cue Test, and Cue Test) as the within-subjects factor and groups (48 h and 6 weeks) as the between-subjects factor (Table 1). As shown on Figure 4A, the group of rats tested 6 weeks after training significantly increased their percentage of freezing during the baseline of the Cue Test session (BL Cue Test; first 3 min of the session) and during the 10 min Cue Test, as compared to BL training (i.e., prior to any exposure to tones and shocks). No analogous difference between BL training and BL Cue was observed for the group of rats tested after 48 h (p > .05). For both groups of rats, the percentage of freezing during the 10 min Cue Test was higher than during the corresponding baseline period of that same session (BL Cue Test), which suggests a retrieval effect. No differences were observed between the groups of rats on percentage of freezing across the different periods (ps > .05).

Figure 4B shows a comparison of mean freezing time (in seconds) specifically during the 10 s tone presentations across Training (tone-shock pairings) and Cue Test (only tone presentations). Only rats tested 6 weeks after training significantly increased the amount of time freezing during the cue.

Lastly, as shown on Figure 4C, only the group of rats tested 48 h after training significantly decreased the freezing time during the ITIs from the Training session to the Cue Test. No differences in freezing time during the ITIs were observed across the two groups of rats (ps >.05).

Figure 4
Figure 4: Effects of an extended cued fear conditioning protocol on freezing response during the cue test.
Data are shown as the mean (bars) and the SEM (error bars) of the freezing response. (A) shows the percentage of freezing of each group of subjects (testing 48 h after training; testing 6 weeks after training) during the first 3 min of the training session (BL, baseline), the first 3 min of the cue test session (BL Cue) and the 10 min of the cue test (Cue Test); a = different from Cue test after 48 h (mean diffBLTraining-Cue48h = 32.84; SE = 10.25; p < .05; Cohen´s d = 1.52); b = different from BL Cue Test (mean diffBLCue-BL6Weeks = 33.98; SE = 8.36; p < .05; Cohen´s d = 1.59) and Cue Test (mean diffCue-BL6Weeks = 55.86; SE = 10.25; p < .05; Cohen´s d = 2.47); c = different from Cue Test after 48 h (mean diffBLCue-Cue48h = 18.99; SE = 5.17; p < .05; Cohen´s d = .67); d = different from Cue Test after 6 weeks (mean diffBLCue-Cue6Weeks = 21.87; SE = 5.17; p < .05; Cohen´s d = .88). (B) shows the average freezing time (in seconds) during cue (tone) of each group of subjects during Training and the Cue Test; * = different from 6 weeks during test period of Cue Test (mean diffTraining-Cue6Weeks = -3.14; SE = 1.37; p < .05; Cohen´s d = 1.64). (C) shows the average freezing time (in seconds) during the intertrial intervals (ITI) of the Training session (10 tone-shock pairings) and the Cue Test (10 tone-only presentations) across the two groups of rats (48 h and 6 weeks); *** = different from Training for group of rats tested 48 h after training (mean diffTraining-Cue48h = 506.16; SE = 95.08; p < .001; Cohen´s d = 2.48). Please click here to view a larger version of this figure.

Discussion

The present extended fear-conditioning protocol is an efficient and valid approach to assess emotional memory across short (48 h) and long-term periods (6 weeks). Thus, the protocol allows to study overtraining and fear incubation phenomena in rats. Among the different advantages of this protocol are the following. It offers two types of memory tests, namely context and cue, that allow to identify the differential effect of two delays (48 h and 6 weeks) across context and cue manipulations. Second, the task entails a single 28 min training session, which in turn produces long-term effects that extend by several weeks. This advantage is remarkable, considering that some versions of extended fear conditioning need at least 100 shocks across 10 sessions of training11. Third, the protocol offers several measurement alternatives, which are calculated automatically. In addition, there is mounting pharmacological, physiological, and anatomical evidence that supports the validity of this paradigm for assessing emotional memory phenomena15,16.

Compared to other fear-conditioning paradigms with brief training sessions (i.e., few trials), extended protocols that result in overtraining effects have received less attention. However, extended fear-conditioning tasks have been key to the understanding of fear incubation’s underlying behavioral and neurobiological processes, including its relationship with other psychological phenomena (e.g., delayed-onset post-traumatic stress disorder)11,12,13. The present fear-conditioning protocol reliably produces fear incubation. This is demonstrated with higher freezing times and lower motion indexes in animals assessed 6 weeks after training, as compared to animals tested 48 h after training. In addition, this effect could be observed differentially in each of the types of test; specifically, longer freezing episodes during the context test 6 weeks after training and increments in freezing during cue presentations 6 weeks after training. Related to this latter effect (i.e., increments in freezing during cue presentations 6 weeks after training), it seems that the novelty of the experimental situation (i.e., new context) can be discarded, since baseline freezing levels during that same session were significantly lower than during the subsequent cue presentations.

Although a trend towards fear learning was evident in both groups (i.e., differences between 3 min baseline and training), animals that were tested after 48 h (context) and 72 h (cue) did not exhibit significant differences in freezing level during both tests. This can be considered a limitation of the protocol, which seems the result of high behavioral variability in the 48 h group (see Figure 2C). A methodological change that can be implemented in order to reduce the variability and improve the procedure is to carry out the context and cue test 24 h after training, which is common in some fear conditioning procedures.

The present protocol could be applied in clinical research23. The strong memory trace and incubation effect that result from its implementation may allow to test the effects of medications regularly used for treatment of psychological and psychiatric pathologies (e.g., anxiolytic or mood regulators treatments24) on emotional memory phenomena (e.g., fear extinction)25,26,27. The protocol thus could allow to measure the influence of medications on the memory trace across different time frames, including biological correlates such as neurotransmitters and molecules related to memory maintenance28,29. The protocol could also be of relevance for research with a translational perspective, which has proposed that fear paradigms could be useful to test preclinical models of behavioral therapies30 and comparative studies on fear across species21,22. Lastly, from a neurobiological view, the present protocol is a robust model to study brain mechanisms, communications between structures, networks or neuronal ensembles involved in long-term acquisition, consolidation and storage of emotional memory, or effects of incubation during development32.

Some other aspects of the protocol are worth discussing. Food deprivation was used throughout the experiment. This decision was adopted because other behavioral tests based on food rewards (e.g., operant or instrumental techniques)33,34,35 can be integrated with minimum changes, making the present protocol a more versatile technique. For instance, we have successfully integrated this protocol with wheel-based exercise protocols and T-maze memory tasks. Another aspect is related to the group size (n=6) implemented in this protocol. Though it was a relatively small sample, and larger samples are certainly recommended, the size of the incubation effect compensates for this limitation (see Table 1). This could be considered an advantage of this protocol, especially regarding animal committees’ recommendations based on the reduction principle. A limitation of the protocol was that minimal or no exposure to footshocks and time-course of fear incubation were not evaluated. An additional control group with the before conditions could increase the rigor of the experimental design.

Final recommendations for the best implementation and results of this protocol include correct cleaning of the experimental chamber, especially the grid floor, calibration of the shock intensities prior to training each subject (e.g., feces and urine often reduce the reliability of the shock intensity across different areas of the chamber) and freezing detection system calibration (reliability of the freezing measures depends on proper setting of motion threshold and minimal freezing duration).

This protocol could be tested with other strains of rats or other rodents (e.g., mice or Mongolian gerbils), broadening the scope of applications. In those cases, it is important to adjust shock intensity, and motion and duration thresholds. The shock intensity used in fear conditioning protocols with mice typically ranges 0.4 mA to 1.5 mA, with 0.75 mA an often reported effective intensity16,36,37,38 and 1.5 mA the highest reported intensity39. The Mongolian gerbil is a rodent model less frequently chosen for fear conditioning research; however, Mongolian gerbils have been successfully used to model circadian rhythms in mammals40. Accordingly, the current protocol could be implemented to study potential relations between circadian rhythms and emotional memory, both of them relevant in pathologies such as depression, anxiety or alteration of mood41,42. In the case of gerbils, an effective shock intensity range for this and analogous aversive conditioning protocols is between 1.0 and 4.0 mA43,44,45,46. Lastly, it is important to note that motion and duration thresholds should be adjusted depending on the species chosen47. These thresholds are the limits established on the movement tracking software, above which the animal behavior is registered as movement and below which the software registers freezing. In aversive conditioning studies with mice and gerbils, effective motion and duration thresholds reported have been 25 and 30 fps (i.e., minimum 1 s immobility), respectively30,35.

To ensure adequate control of aversive stimulation (footshocks), all sectors of the grid floor must deliver the same intensity. It is recommended to calibrate the intensity of the shock in three sectors of the grid floor to verify that it is consistent. This prevents animals from learning to reduce exposure to the shocks by moving to a place in the box that emits a lower intensity. If the calibration shows that the metal grid is not delivering the same intensity in all sectors, remove the grid from the floor, clean the rods, and replace the grid in the chamber. The grid floor must be properly inserted into the chamber to ensure the best electric transmission from the aversive stimulation device to the grid floor.

The focus and aperture of the freezing detection system camera is calibrated by the manufacturer. However, if additional calibration is required, loosen the setscrew on the focus ring, adjust until a clear image is achieved and then tighten the setscrew on the focus ring. The manufacturer recommends locking the lens opening in the maximum open position. To achieve this setting, make sure that the white point of the opening ring is aligned with the number 1.4 on the lens barrel. It is recommended to consult the manufacturer's manual. Note that if the focus of the camera was adjusted, calibration of the camera using the corresponding software must also occur. Camera calibration requires adjustment of the brightness, gain, and shutter. It is recommended to consult the manufacturer's manual for precise instructions on the camera calibration process.

In conclusion, the protocol allows to test emotional memory across short and long-term periods and produces long-term fear incubation. This fear-incubation effect is generated via a single-session overtraining, which shows effects 6 weeks later in context and cue tests. This suggests a strong emotional memory trace. The current protocol is an efficient and valid approach to explore the components of emotional memory in rats.

Divulgazioni

The authors have nothing to disclose.

Acknowledgements

Financial support for this research was provided by Fundación Universitaria Konrad Lorenz – grant number 9IN15151. The authors would like to thank the Communications Department at Konrad Lorenz University for their help with recording and editing the video, in particular Natalia Rivera and Andrés Serrano (Producers). Also, Nicole Pfaller-Sadovsky and Lucia Medina for their comments on the manuscript, and Johanna Barrero, Dean at Corporacion Universitaria Iberoamericana, for institutional collaboration. The authors have no conflicts of interest.

Materials

Acetic acid (ethanoic acid) https://pubchem.ncbi.nlm.nih.gov/compound/acetic_acid
Aversive Stimulation Current Package MED Associates Inc ENV-420 https://www.med-associates.com/product/aversive-stimulation-current-test-package/
Contextual test protocol.pro https://osf.io/4nkfq/?view_only=0640852a88544b239549462f9c21175b.
Cue test protocol.pro https://osf.io/4nkfq/?view_only=0640852a88544b239549462f9c21175b.
Curved Wall Insert MED Associates Inc VFC-008-CWI https://www.med-associates.com/product/curved-wall-insert/
Data processing.zip https://osf.io/4nkfq/?view_only=0640852a88544b239549462f9c21175b.
NIR/White Light Control Box MED Associates Inc NIR-100
Pellets BioServ F0165 http://www.bio-serv.com/pdf/F0165.pdf
Quick Change Floor/Pan Unit for Mouse MED Associates Inc ENV-005FPU-M https://www.med-associates.com/product/quick-change-floorpan-unit-for-mouse/
Small Tabletop Cabinet and Power Supply MED Associates Inc SG-6080D https://www.med-associates.com/product/small-tabletop-cabinet-and-power-supply-120v-60-hz/
Standalone Aversive Stimulator/Scrambler (115 V / 60 Hz) MED Associates Inc ENV-414S https://www.med-associates.com/product/standalone-aversive-stimulatorscrambler-115-v-ac-60-hz/
Standard Fear Conditioning Chamber MED Associates Inc VFC-008 https://www.med-associates.com/product/standard-fear-conditioning-chamber/
Training protocol VFC.pro https://osf.io/4nkfq/?view_only=0640852a88544b239549462f9c21175b.
Video Fear Conditioning Package for Rat MED Associates Inc MED-VFC-SCT-R https://www.med-associates.com/product/nir-video-fear-conditioning-system-for-rat/

Riferimenti

  1. Frankland, P. W., Bontempi, B. The organization of recent and remote memories. Nature Reviews Neuroscience. 6 (2), 119-130 (2005).
  2. Suzuki, A., Mukawa, T., Tsukagoshi, A., Frankland, P. W., Kida, S. Activation of LVGCCs and CB1 receptors required for destabilization of reactivated contextual fear memories. Learning & Memory. 15 (6), 426-433 (2008).
  3. Hermans, E. J., et al. How the amygdala affects emotional memory by altering brain network properties. Neurobiology of Learning and Memory. 112, 2-16 (2014).
  4. Moryś, J., Berdel, B., Jagalska-Majewska, H., ŁUczyńSka, A. The basolateral amygdaloid complex -its development, morphology and functions. Folia Morphologica. 58 (3), 29-46 (1998).
  5. LeDoux, J. E. Emotional memory systems in the brain. Behavioural Brain Research. 58 (1-2), 69-79 (1993).
  6. Labar, K. S. Beyond fear: Emotional memory mechanisms in the human brain. Current Directions in Psychological Science. 16 (4), 173-177 (2007).
  7. Izquierdo, I., Furini, C. R. G., Myskiw, J. C. Fear Memory. Physiological Reviews. 96 (2), 695-750 (2016).
  8. Maren, S. Overtraining Does Not Mitigate Contextual Fear Conditioning Deficits Produced by Neurotoxic Lesions of the Basolateral Amygdala. The Journal of Neuroscience. 18 (8), 3097-3097 (1998).
  9. Pickens, C. L., Golden, S. A., Nair, S. G. Incubation of fear. Current Protocols in Neuroscience. 64, (2013).
  10. Morrow, J. D., Saunders, B. T., Maren, S., Robinson, T. E. Sign-tracking to an appetitive cue predicts incubation of conditioned fear in rats. Behavioural Brain Research. 276, 59-66 (2015).
  11. Pickens, C. L., Golden, S. A., Adams-Deutsch, T., Nair, S. G., Shaham, Y. Long-lasting incubation of conditioned fear in rats. Biological Psychiatry. 65 (10), 881-886 (2009).
  12. Schaap, M. W. H., et al. Nociception and Conditioned Fear in Rats: Strains Matter. PLoS ONE. 8 (12), 83339 (2013).
  13. Shoji, H., Takao, K., Hattori, S., Miyakawa, T. Contextual and Cued Fear Conditioning Test Using a Video Analyzing System in Mice. Journal of Visualized Experiments. (85), e50871 (2014).
  14. Patel, T. P., et al. An open-source toolbox for automated phenotyping of mice in behavioral tasks. Frontiers in Behavioral Neuroscience. 8, 349 (2014).
  15. Kabra, M., Robie, A. A., Rivera-Alba, M., Branson, S., Branson, K. JAABA: Interactive machine learning for automatic annotation of animal behavior. Nature Methods. 10 (1), 64-67 (2013).
  16. Anagnostaras, S. G. Automated assessment of Pavlovian conditioned freezing and shock reactivity in mice using the VideoFreeze system. Frontiers in Behavioral Neuroscience. 4 (58), (2010).
  17. Moyer, J. R., Brown, T. H. Impaired Trace and Contextual Fear Conditioning in Aged Rats. Behavioral Neuroscience. 120 (3), 612-624 (2006).
  18. Schuette, S. R., Hobson, S. Conditioned contextual fear memory to assess natural forgetting and cognitive enhancement in rats. Journal of Biological Methods. 5 (3), 99 (2018).
  19. Chang, C. H., et al. Fear extinction in rodents. Current Protocols in Neuroscience. , (2009).
  20. Pickens, C. L., Golden, S. A., Nair, S. G. Incubation of fear. Current Protocols in Neuroscience. 64, 1-18 (2013).
  21. Izquierdo, I., Furini, C. R. G., Myskiw, J. C. Fear Memory. Physiological Reviews. 96 (2), 695-750 (2016).
  22. Vetere, G., et al. Chemogenetic Interrogation of a Brain-wide Fear Memory Network in Mice Article Chemogenetic Interrogation of a Brain-wide Fear Memory Network in Mice. Neuron. 94 (2), 363-374 (2017).
  23. Koob, G. F., Zimmer, A. Chapter 9 – Animal models of psychiatric disorders. Neurobiology of Psychiatric Disorders. 106, 137-166 (2012).
  24. Bourin, M. Animal models for screening anxiolytic-like drugs: a perspective. Dialogues in clinical neuroscience. 17 (3), 295-303 (2015).
  25. Murray, S. B., et al. Fear as a translational mechanism in the psychopathology of anorexia nervosa. Neuroscience & Biobehavioral Reviews. 95, 383-395 (2018).
  26. Pamplona, F. A., et al. Prolonged fear incubation leads to generalized avoidance behavior in mice. Journal of Psychiatric Research. 45 (3), 354-360 (2011).
  27. Török, B., Sipos, E., Pivac, N., Zelena, D. Modelling posttraumatic stress disorders in animals. Progress in Neuro-Psychopharmacology and Biological Psychiatry. 90, 117-133 (2019).
  28. Bhakta, A., Gavini, K., Yang, E., Lyman-Henley, L., Parameshwaran, K. Chronic traumatic stress impairs memory in mice: Potential roles of acetylcholine, neuroinflammation and corticotropin releasing factor expression in the hippocampus. Behavioural Brain Research. 335, 32-40 (2017).
  29. Uniyal, A., et al. Pharmacological rewriting of fear memories: A beacon for post-traumatic stress disorder. European Journal of Pharmacology. , 172824 (2019).
  30. Barad, M. Fear extinction in rodents: basic insight to clinical promise. Current Opinion in Neurobiology. 15 (6), 710-715 (2005).
  31. Haaker, J., et al. Making translation work: Harmonizing cross-species methodology in the behavioural neuroscience of Pavlovian fear conditioning. Neuroscience & Biobehavioral Reviews. 107, 329-345 (2019).
  32. Heroux, N. A., Horgan, C. J., Pinizzotto, C. C., Rosen, J. B., Stanton, M. E. Medial prefrontal and ventral hippocampal contributions to incidental context learning and memory in adolescent rats. Neurobiology of Learning and Memory. 166, 107091 (2019).
  33. Rossi, M. A., Yin, H. H. Methods for Studying Habitual Behavior in Mice. Current Protocols in Neuroscience. 60 (1), 8-29 (2012).
  34. Brady, A. M., Floresco, S. B. Operant Procedures for Assessing Behavioral Flexibility in Rats. Journal of Visualized Experiments. (96), (2015).
  35. Zoccolan, D., Di Filippo, A. Methodological Approaches to the Behavioural Investigation of Visual Perception in Rodents. Handbook of Behavioral Neuroscience. , (2018).
  36. Lguensat, A., Bentefour, Y., Bennis, M., Ba-M’hamed, S., Garcia, R. Susceptibility and Resilience to PTSD-Like Symptoms in Mice Are Associated with Opposite Dendritic Changes in the Prelimbic and Infralimbic Cortices Following Trauma. Neuroscienze. 418, 166-176 (2019).
  37. Li, Q., et al. N-Acetyl Serotonin Protects Neural Progenitor Cells Against Oxidative Stress-Induced Apoptosis and Improves Neurogenesis in Adult Mouse Hippocampus Following Traumatic Brain Injury. Journal of Molecular Neuroscience. 67 (4), 574-588 (2019).
  38. Pantoni, M. M., Carmack, S. A., Hammam, L., Anagnostaras, S. G. Dopamine and norepinephrine transporter inhibition for long-term fear memory enhancement. Behavioural Brain Research. 378 (112266), 112266 (2020).
  39. Smith, K. L., et al. Microglial cell hyper-ramification and neuronal dendritic spine loss in the hippocampus and medial prefrontal cortex in a mouse model of PTSD. Brain, Behavior, and Immunity. 80, 889-899 (2019).
  40. Liu, X., Zheng, X., Liu, Y., Du, X., Chen, Z. Effects of adaptation to handling on the circadian rhythmicity of blood solutes in Mongolian gerbils. Animal Models and Experimental. 2 (2), 127-131 (2019).
  41. Landgraf, D., McCarthy, M. J., Welsh, D. K. The role of the circadian clock in animal models of mood disorders. Behavioral Neuroscience. 128 (3), 344-359 (2014).
  42. Refinetti, R., Kenagy, G. J. Diurnally active rodents for laboratory research. Laboratory annimals. 52 (6), 577-587 (2018).
  43. Hurtado-Parrado, C., et al. Assessing Mongolian gerbil emotional behavior: effects of two shock intensities and response-independent shocks during an extended inhibitory-avoidance task. PeerJ. 5, (2017).
  44. Frey, P., Eng, S., Gavinf, W. Conditioned suppression in the gerbil. Behavior Research Methods & Instrumentation. 4 (5), 245-249 (1972).
  45. Woolley, M. L., Haman, M., Higgins, G. A., Ballard, T. M. Investigating the effect of bilateral amygdala lesions on fear conditioning and social interaction in the male Mongolian gerbil. Brain Research. 1078 (1), 151-158 (2006).
  46. Ballard, T. M., Sänger, S., Higgins, G. a Inhibition of shock-induced foot tapping behaviour in the gerbil by a tachykinin NK1 receptor antagonist. European Journal of Pharmacology. 412 (3), 255-264 (2001).
  47. Luyten, L., Schroyens, N., Hermans, D., Beckers, T. Parameter optimization for automated behavior assessment: plug-and-play or trial-and-error. Frontiers in Behavioral Neuroscience. 8 (28), (2014).

Play Video

Citazione di questo articolo
Acevedo-Triana, C., Rico, J. L., Ortega, L. A., Cardenas, M. A. N., Cardenas, F. P., Rojas, M. J., Forigua-Vargas, J. C., Cifuentes, J., Hurtado-Parrado, C. Fear Incubation Using an Extended Fear-Conditioning Protocol for Rats. J. Vis. Exp. (162), e60537, doi:10.3791/60537 (2020).

View Video