How to Calculate Spontaneous Alternation in a Y-Maze

1. What spontaneous alternation measures

Spontaneous alternation is a widely used Y-Maze measure of an animal's tendency to visit each of the maze's three arms before returning to one it has already visited, rather than repeatedly re-entering the same one or two arms. It is calculated purely from the chronological order of confirmed arm entries -- no other tracked measure (dwell time, distance, trajectory) feeds into it.

It is reported as a single percentage per experiment, alongside the entry counts it was computed from.

2. The RodentMetrics formula

RodentMetrics scores spontaneous alternation from the confirmed, chronological arm-entry sequence using overlapping windows of three consecutive entries ("triads"). This is the standard, widely used spontaneous-alternation formula -- RodentMetrics does not use a modified or proprietary variant of it.

Term	Definition
Total entries	The number of confirmed arm entries in the session.
Total triads	Total entries − 2 (the count of overlapping, consecutive 3-entry windows). Undefined when fewer than 3 entries are confirmed.
Correct triads	The number of those triads whose three entries are into three distinct arms.
Spontaneous alternation (%)	Correct triads ÷ total triads × 100.

Undefined, not zero, below three entries

A session with fewer than three confirmed arm entries has not produced a single scoreable triad. RodentMetrics reports the percentage as undefined in that case rather than as 0% -- 0% would incorrectly imply a triad was scored and failed.

3. What counts as a confirmed arm entry

An arm entry is not simply "the tracked point is inside an arm." RodentMetrics requires a directional, temporally confirmed crossing:

the animal must first be confirmed in the center junction for a minimum dwell before a crossing can begin;
its tracked position must then cross a virtual boundary a fixed fraction of the way into the arm, moving from center toward the arm;
it must remain past that boundary continuously, for a minimum duration, before the entry is confirmed.

Missing or non-consecutive tracking frames break an in-progress entry rather than being bridged or interpolated. A direct transition between two arms is never accepted as a valid step in the sequence unless an actual passage through the center junction was observed in between -- it is either bridged through that observed center run, or logged separately as an invalid transition and excluded from the confirmed entry sequence entirely.

4. Worked example

Take the illustrative, synthetic arm-entry sequence A → B → C → A → C → B (six confirmed entries; not data from a real recording). With six entries there are four overlapping triads:

Triad	Entries	Distinct arms	Correct alternation?
1	A, B, C	3	Yes
2	B, C, A	3	Yes
3	C, A, C	2 (A, C)	No
4	A, C, B	3	Yes

3 of the 4 triads are correct alternations, so RodentMetrics would report 75% (3 ÷ 4 × 100) for this sequence, alongside total entries (6), total triads (4) and correct triads (3).

5. Repeated entries and other edge cases

The formula applies the same triad rule regardless of whether an arm repeats. Take the synthetic sequence A → A → B → C (the animal re-enters arm A immediately after leaving it, then goes to B and C):

Triad	Entries	Distinct arms	Correct alternation?
1	A, A, B	2 (A, B)	No
2	A, B, C	3	Yes

1 of the 2 triads is correct, so this sequence scores 50%. No special-case handling is applied to a repeated arm -- it is simply not three distinct arms in whichever triad it appears in.

Two related behaviors worth knowing about before interpreting a repeated-entry sequence:

A same-arm return via only a very brief center touch is treated as tracking jitter and suppressed -- it is not logged as a separate exit-then-reentry, so it will not appear as a duplicate consecutive entry in the confirmed sequence.
A genuinely confirmed dwell in the center junction followed by re-entering the same arm is recorded as an ordinary, separate entry, exactly as in the example above.
Sessions with fewer than three confirmed entries report the percentage as undefined (see Section 2), not as 0%.

6. What this metric does not establish

Spontaneous alternation reports a proportion of triads that met a geometric distinct-arm criterion. RodentMetrics presents it in that descriptive form and avoids stronger claims. In particular, this percentage alone does not:

diagnose or measure a cognitive deficit;
prove or disprove a memory impairment;
validate an anxiety-like or emotional state;
constitute a clinically validated outcome measure.

The metric represents one behavioral pattern derived from confirmed arm-entry order. It should be interpreted alongside total entry count, tracking/quality information for the session, and the experimental context and protocol under which it was collected -- not read in isolation as a conclusion about the animal.

7. What to report

When publishing or exporting Y-Maze spontaneous alternation results, RodentMetrics recommends reporting the parameters it actually computes and exports, so a reader can evaluate the number rather than take it on faith:

Total confirmed arm entries for the session
Total eligible triads and correct triads (not just the final percentage)
The spontaneous alternation percentage, or that it was undefined if fewer than three entries were confirmed
Per-arm entry counts
Tracking/quality information for the session (e.g. classified behavioral coverage)
The RodentMetrics software version the analysis was run with