Measures
Direct path measures
A straight line to the platform is the clearest sign of a place memory.
Direct path measures ask a simple question of each trial: did the animal swim more or less straight to the platform? They're a quick way to tell precise spatial searching from luck or habit.
Why it matters
Direct Path tells you when a subject swam straight to the target at the start of a trial, a strong sign it knew where to go.
More detail
Six classifications identify initial goal-directed navigation to the target or its close-encounter zone, directly, following orientation, or from the poolside. They preserve information about the early trajectory that measures summarising a larger portion of the trial may not show.
A path counts as direct when actual path divided by optimal path is 1.2 or less. Failing to meet a classification does not establish impaired learning, memory or localisation (Direct Path Measures technical note, doi:10.5281/zenodo.21870806).
Direct Path Measures
Direct Path classifications say whether a trial began with an accurate approach. There are two groups of three: Direct Path to Target, …following Orientation and …from Poolside, and the same three for Close Encounter. Within each group only one is reported.
More detail
The order of reporting is from the start first, then following Orientation, then from Poolside. A path counts as direct when actual path / shortest path is no more than 1.2. Direct Path complements latency and path length rather than replacing them.
Classifications consider only the first arrival and are evaluated only when the trial starts outside the destination; they are not produced for time slices. A trial with Direct Path to Target, Direct Path to Close Encounter or a following-Orientation form is not also identified as Chaining. Cite the HVS Image Direct Path Measures documentation (doi:10.5281/zenodo.21870806).
Questions people ask
What are the six Direct Path classifications and which one is reported?
A Direct Path is an approach to the target that is close to a straight line: the path actually travelled is no more than 1.2 times the shortest possible path. HVS Image has six classifications in two groups of three: Direct Path to Target, …following Orientation and …from Poolside, and the same three for Close Encounter (the zone around the target). Within each group only one is reported: from the start first, then following Orientation, then from Poolside.
More detail
A Direct Path is an efficient, near-straight approach to a destination, defined in HVS Image as actual path divided by the straight-line distance being no more than 1.2. Direct Path to Target: the path from trial start to first entry into the target, divided by the straight-line distance to the target's nearest edge, is ≤ 1.2. Following Orientation: the earliest point from which the rest of the path is direct begins by 0.25 × pool radius of travelled path, or by 0.50 × radius if the Heading Angle is ≤ 15°. From Poolside: the direct approach begins by the late orientation point, from within the thigmotaxis zone (default outer 20% of radius), without having progressed more than one-third of the radius inwards beforehand, and the target is reached during the first visit to the close encounter zone. Close Encounter versions use the close encounter zone as destination and require that the target is not reached during the first visit to that zone. A trial can carry both a target and a close encounter classification.
Going deeper
Classifications consider only the first arrival and are evaluated only when the trial starts outside the destination. They describe the whole trial's initial approach and are not produced for time slices. A trial with Direct Path to Target, Direct Path to Close Encounter or a following-Orientation form is not also identified as Chaining. Cite DOI 10.5281/zenodo.21870806.
Which analysis settings change the Direct Path results, and what should I report?
Three settings matter: Counter diam x (close encounter size), ThigRatio (poolside zone) and Heading distance (where Heading Angle is measured). Keep them constant across a study and report them. Check the exported Heading range rather than assuming a value.
More detail
A Direct Path is a near-straight first approach to the target, with an actual path no more than 1.2 times the direct distance. Counter diam x defaults to 2.0 (close encounter radius twice the target radius). Heading distance × pool diameter sets the heading point; the program Default gives 0.4 (0.8 × radius), but the shipped configuration file has contained 0.2 (0.4 × radius), so the value in use is exported as Heading range. The 1.2 ratio, the 0.25R and 0.50R limits, 15° and the 2R/3 poolside limit are fixed criteria. Report the build and these settings with re-analysed data.
Going deeper
2024-era heading used path length = 2.5 × pool radius × the Advanced value (0.2 default, i.e. half the radius); to match the 2026 default of 0.4 × diameter in the 2024 version, set 0.32. Where tracking is briefly lost, the gap is bridged by a straight line, so apparent directness may partly reflect unobserved movement.
Does 'from Poolside' mean the animal followed the wall?
No. 'From Poolside' describes where the direct approach began (near the wall), not continuous wall-following. Sustained wall-following is reported separately as Thigmotaxis.
More detail
No: 'from Poolside' is a Direct Path classification for where the direct approach began, near the wall, and it does not mean the subject followed the wall. Thigmotaxis, by contrast, is sustained wall-hugging, reported separately as % time and % path in the thigmotaxis band. Brief poolside movement should not necessarily be read as thigmotaxis or failure to locate the target. Persistent peripheral movement is common early in training and may reflect anxiety- or stress-related behaviour or non-spatial factors. Some web pages call this class 'after Leaving the Wall'; the current definition uses 'from Poolside'.
Going deeper
A trial may show a Poolside Direct Path followed by Chaining, suggesting a change of strategy after the initial approach.
Two related tests are explained on hvsimage.com: the corridor test and the cone test.
Spotted a mistake, or know a better source? Tell us and we'll fix it and say so on our corrections page.