What Is the Elevated Plus Maze?
The Elevated Plus Maze (EPM) was developed by Pellow et al. (1985) and rapidly became one of the most widely used preclinical tests for anxiety-like behavior. The apparatus has a plus (+) shape with four arms: two open arms (typically 30-50 cm long, 5-10 cm wide, no walls) and two closed arms (same dimensions but enclosed by 15-40 cm high walls), connected by a central square platform. The entire maze is elevated 40-70 cm above the floor.
The test exploits a natural approach-avoidance conflict: rodents are motivated to explore novel environments but are also innately averse to open, elevated, brightly lit spaces. Anxious animals spend more time in the protected closed arms, while animals treated with anxiolytic compounds (e.g., diazepam, chlordiazepoxide) show increased open arm exploration. A single 5-minute trial is the standard protocol — the EPM is notable for not requiring training or food/water deprivation.
The EPM has been pharmacologically validated across species (mice, rats) and is sensitive to benzodiazepines, SSRIs (with chronic administration), 5-HT1A agonists, and various genetic manipulations. It is used extensively in anxiety disorder research, drug screening, and phenotyping of transgenic/knockout mouse lines.
EPM Metrics Explained
The EPM produces several complementary behavioral metrics, each capturing a different aspect of anxiety-like behavior and locomotion.
% Open Arm Time = (time in open arms / total time) x 100. This is the primary anxiety measure. Lower values indicate higher anxiety. It is the most sensitive EPM metric to anxiolytic drugs and the most commonly reported outcome.
% Open Arm Entries = (open arm entries / total entries) x 100. This normalizes open arm exploration for overall activity level. It is less affected by locomotor confounds than raw open arm entries and is recommended as a complementary measure to % open arm time.
Anxiety Index = 1 - (open arm time / total time). A composite score ranging from 0 (no anxiety) to 1 (maximum anxiety). It provides an intuitive single-number summary that inverts the % open arm time scale so that higher values = more anxiety.
Total Entries = open arm entries + closed arm entries. A validated proxy for general locomotor activity. Essential for ruling out sedation or hyperactivity as confounds. If total entries differ significantly between groups, open arm metrics must be interpreted cautiously.
Center Time. Time on the center platform reflects risk assessment behavior — the decision point between approach (open arm) and avoidance (closed arm). Elevated center time may indicate increased conflict or hesitation.
Best Practices and Common Pitfalls
The EPM is deceptively simple to run but requires careful attention to procedural details that can dramatically affect results.
Lighting: Open arm illumination should be standardized (typically 200-300 lux). Brighter light increases anxiety; dimmer light reduces it. Measure and report light levels on both open and closed arms.
One-trial tolerance: Unlike most behavioral tests, the EPM should only be run once per animal. Repeated testing produces a robust decrease in open arm exploration that is resistant to anxiolytic drugs (File et al. 1990). This "one-trial tolerance" means the EPM cannot be used for within-subject pre/post designs.
Starting position: Always place the animal in the center facing a closed arm (or consistently facing the same direction). Starting position affects first-choice behavior, which in turn influences all subsequent exploration.
Cleaning: Clean the maze between animals with 70% ethanol or a mild detergent and allow it to dry. Residual odor cues from a previous animal (especially alarm pheromones from stressed animals) can alter behavior.
Time of day: Test at the same time daily. Circadian variation in anxiety-like behavior is well documented in rodents. Most labs test during the light phase, 2-6 hours after lights-on.
Exclusion criteria: Pre-define exclusion criteria before testing. Common reasons include: animal falls off the maze, animal remains immobile for the entire session, or total entries = 0 (suggesting the animal is not exploring at all).
