Recommended Free Tools
Female green swordtails have preferred larger male images in controlled visual tests, but the evidence does not show that females universally choose a longer sword ornament. In the key experiments, enlarging a swordless male could erase the apparent advantage of a sworded male, indicating that overall apparent size can drive the response. A 2023 study further found that the proportional size gap between two males predicted preference better than their raw size difference.
What “preference” means in these experiments
In behavioral studies, “preference” usually means where a female spends time when two visual stimuli are available at once. The fish is placed in a choice tank with a stimulus on each side. Researchers divide the tank into left, centre and right zones, then record the time the female associates with each stimulus.
As an Amazon Associate I earn from qualifying purchases.
The common calculation is:
preference difference = association time beside the larger stimulus − association time beside the smaller stimulus
A positive value means more time near the larger image; a value near zero means little measurable difference. This is a behavioral proxy for attraction or choice under the test conditions. It is not a direct count of courtship, copulation, fertilization or offspring.
#1 Best Overall
Why a sword can make a male look larger
Male green swordtails carry an elongated lower tail-fin ray known as the sword. Removing that ornament from an image reduces the male’s total visible length, even if the body itself is unchanged. A female viewing the whole silhouette may therefore respond to the animal’s apparent overall size rather than evaluating sword length as an independent trait.
This distinction matters because a test that compares an intact, sworded male with a swordless male can make the sword appear to be the decisive signal. To determine whether the ornament itself is responsible, researchers must control the rest of the visual impression, including the total projected size.
What the 1998 video experiment tested
Manipulating the courtship image
Rosenthal and Evans used computer-altered video sequences of courting Xiphophorus helleri males. Their comparisons included intact males, males from which the sword portion had been removed, and isolated sword stimuli. Because the researchers could edit the same recorded material, they could change the ornament while holding much of the male’s behavior and appearance constant.
The Tool Desk
Outbyte PC Repair FREERepair Windows errors before they cause bigger problemsFix Now →Outbyte Driver Updater FREEScan for outdated or missing drivers - takes under a minuteDriver Scan →The critical size-compensation control
The decisive control enlarged the swordless male image to compensate for the length removed with the sword. Before that compensation, females preferred the image of the sworded male. Once the swordless image was enlarged so that its apparent total length matched the relevant visual extent, the preference for the sworded image disappeared.
The result does not prove that sword length is irrelevant. It shows that, in this manipulation, apparent overall size was sufficient to explain the earlier response. The study’s conclusion was consistent with a general bias toward large apparent mates rather than a response to a sword-specific character.
What this design can and cannot isolate
- It can test: whether changing sword presence while controlling apparent size changes the response.
- It can suggest: that a visual preference attributed to the sword may actually be mediated by total apparent length.
- It cannot establish: that every population of green swordtails, or every Xiphophorus species, uses the same rule in natural courtship.
How the 2023 study separated absolute and proportional size
Standardized animated males
Caves and colleagues used two-dimensional animations of courting green swordtail males, identified in the paper as Xiphophorus hellerii. They created scaled versions of the same male image. Traits other than size were therefore held constant as far as the animation allowed, reducing the chance that color, posture or individual appearance would explain a choice.
Choice-tank procedure
The study tested 23 females with 11 stimulus pairs and two presentations per pair. During a viewing period, the female was held in the centre while both animations were visible. The fish was then released for three minutes of free movement while the animations continued. Time in the left, centre and right zones was tracked, and the two association times were used to calculate the preference difference.
Do these 3 things before closing this tab:
1Repair Windows errors before they cause bigger problems2Scan for outdated or missing drivers - takes under a minute3Clear out junk files and repair common Windows errorsFor a repeat presentation, the researchers reversed which side showed the larger animation. This counterbalanced left-right placement so that a persistent side bias would be less likely to masquerade as a size preference.
Rank #3
Two competing definitions of “difference”
An absolute-difference analysis asks whether a fixed area gap matters. For example, a 20-unit area difference is treated as equally meaningful whether the stimuli are 100 and 120 units or 200 and 220 units.
A proportional-difference analysis asks how large the gap is relative to the stimuli themselves. The same 20-unit gap is a 20% increase in the first pair but only a 10% increase in the second. The 2023 experiment deliberately paired stimuli across ranges of absolute and proportional differences so these explanations could be compared.
What the model comparison found
The authors compared mixed-effects models, treating fish identity as a random effect. Candidate explanations used proportional difference, absolute difference, mean stimulus size, or no systematic size effect. The proportional-difference model had the strongest support.
| Candidate model | ΔAIC relative to proportional model | Interpretation |
|---|---|---|
| Proportional difference | 0 (reference model) | Best-supported explanation among the main models |
| Mean size | 10.01 | Much weaker support than proportional difference |
| Absolute difference | 11.63 | Much weaker support than proportional difference |
| Null model | 27.16 | Very little support compared with the proportional model |
As the proportional gap increased, females generally showed a stronger preference for the larger animation. In four of five comparisons where pairs shared the same absolute area difference but differed in proportional difference, the pair with the larger proportional gap produced the stronger preference. Those are experiment-specific comparisons, not a population-wide percentage.
Why relative size can matter more than a raw size gap
A proportional rule is plausible for a visual system that evaluates signals against a reference scale. A fixed difference may be easy to detect when two males are small but less consequential when both are large. Under proportional processing, the visual system responds to how different the two options look in relation to their overall magnitude.
Consider two hypothetical pairs with an identical 20-unit area difference:
| Pair | Smaller stimulus | Larger stimulus | Absolute gap | Proportional gap |
|---|---|---|---|---|
| A | 100 units | 120 units | 20 units | 20% |
| B | 200 units | 220 units | 20 units | 10% |
A model based on absolute difference predicts the same strength for both pairs. A proportional model predicts a stronger response to Pair A. The 2023 results favored the second type of prediction, although the exact visual units and response strength belong to that study’s experimental design.
The “opposite miss” pattern at large sizes
The study also reported that preference nearly disappeared in some pairs where both stimuli were large and their relative difference was small. Its additional analysis favored an “opposite miss” pattern: discrimination at high magnitudes was poorer than a simple proportional-processing rule would predict.
Best Value
This finding should be read narrowly. It does not mean that large males are ignored, nor that females stop preferring larger males. It means that when both alternatives were already large and close together in relative terms, the measured difference in association time could become weak. The pattern was observed in this laboratory experiment with these animations and should not be treated as a universal rule for swordtails.
Independent reader supportYour contribution helps us test, update, and keep practical guides available for everyone.How the two major experiments differ
| Feature | Rosenthal and Evans (1998) | Caves and colleagues (2023) |
|---|---|---|
| Species name used by authors | Xiphophorus helleri | Xiphophorus hellerii |
| Stimulus | Computer-altered video of courting males and sword manipulations | Two-dimensional animations of courting males scaled from the same male image |
| Main question | Whether an apparent sword preference could be explained by total apparent size | Whether absolute difference, proportional difference, mean size or no size effect best predicted preference |
| Key control | Enlarged swordless image to compensate for the removed sword length | Pairs varied in absolute and proportional area differences; side was reversed on repeat trials |
| Behavioral measure | Response to paired visual stimuli | Association time in left, centre and right zones |
| Reported sample and schedule | Not stated in the summary of the 1998 study used here | 23 females, 11 stimulus pairs, two presentations per pair |
| Main result | Sworded-image preference disappeared after size compensation | Proportional difference outperformed absolute difference and other tested models |
How to interpret “larger male preference” responsibly
It is a visual result, not a reproductive outcome
Association time tells you which side of a test arena a female occupied while stimuli played. A female may spend longer near one image without mating with the corresponding male in nature. To establish effects on copulation, fertilization or offspring, researchers would need behavioral or reproductive measurements beyond the visual choice assay.
Animations and video improve control but reduce realism
Playback and animation let researchers alter one feature while keeping others stable. That control is the reason the sword-removal and size-compensation test is informative. At the same time, a screen image is not a live male: movement, three-dimensional shape, chemical cues, social interactions and habitat conditions are simplified or absent.
Quick wins for a faster PC:
Clear out junk files and repair common Windows errorsFree Scan →Scan for outdated or missing drivers - takes under a minuteDriver Scan →Species and populations matter
The highlighted studies concern green swordtails, with the species written as Xiphophorus helleri in the 1998 paper and Xiphophorus hellerii in the 2023 paper. Findings from these experiments should not automatically be extended to every swordtail species, population or ecological setting.
A practical checklist for reading a swordtail preference study
- Identify the stimulus: Was it a live male, video playback or computer-generated animation?
- Ask what was isolated: Did the manipulation change sword length, body size, total projected size or several traits at once?
- Check the choice schedule: Were both stimuli shown simultaneously, or were they presented one after another?
- Find the response variable: Was preference based on association time, a difference between zones, courtship behavior or an eventual reproductive event?
- Separate absolute from proportional contrast: A raw size gap and a percentage gap can lead to different predictions.
- Check the scale of the pair: Preference may depend on mean size as well as the difference between alternatives.
- Look for side-bias controls: Reversing stimulus positions helps distinguish visual choice from a fixed tank-side preference.
- Keep the scope bounded: Note the species, female sample, laboratory conditions and whether the result was replicated outside the original setup.
Bottom line for the evidence
The strongest supported explanation is not that female swordtails universally select the longest sword. In the 1998 manipulation, restoring the apparent length of a swordless male removed the sworded image’s advantage. In the 2023 experiment, the relative gap between two male images predicted association-time preference better than their absolute gap. Together, these studies show how a seemingly ornament-specific choice can reflect general visual processing of apparent size, while leaving open how the preference operates during natural courtship and reproduction.
Quick Recap
Product prices and availability are accurate as of the date/time indicated and are subject to change. Any price and availability information displayed on Amazon at the time of purchase will apply.
