Why are my Neocaridina hiding or swimming around?

Evidence review: August 12, 2026

Describe who moved, where, how, when and for how long before naming a cause. Hiding can occur without fish. Swimming can occur without mating. A sudden colony-wide change deserves investigation even when the movement looks familiar.

Use the anatomy map when a movement depends on a named body region or appendage.

Start with a pattern, not "normal"

Record these seven dimensions:

1. Who: one animal, several, or most of the visible colony?
2. Where: conditioned surfaces, shelter, glass, open water, high-flow zone or waterline?
3. How: grazing, walking, controlled swimming, rapid repeated swimming, stillness, loss of balance or inability to right?
4. When: exact time, light state and time since feeding, maintenance, transfer or another disturbance?
5. How long: seconds, minutes, hours or repeated across days?
6. What changed: food, water, plants, tankmates, chemical, room conditions, flow, air or temperature?
7. Compared with what: this tank at the same time and feeding state on earlier dates?

Do not silently count an unseen shrimp as hidden, dead, escaped or eaten. Unseen remains unresolved until the observation method establishes more.

Hiding is common even without fish

A controlled 72-hour study observed males, non-ovigerous females and ovigerous females every three hours in 23 L tanks with no fish. On average, 82.1% of animals were using shelter at each observation. Shelter use averaged 88.79% by day and 75.44% by night. Ovigerous females sheltered most.¹

Of the four structures offered, Vesicularia moss was used most overall, followed by wood; rock was used least. This supports offering complex shelter. It does not prove that moss reduces stress, determine an ideal hidden percentage, measure juvenile survival or make a fish pairing safe.

Material, shape and structural complexity changed together, so this was not a clean test of complexity or a live-plant effect. Five ten-shrimp aquarium groups represented each sex or reproductive class, and the same groups were observed repeatedly. The paper does not state a repeated-measures or aquarium-level model. Animals were not fed during the 72 hours, and the nighttime observer used a red lamp. Those conditions belong to the result.

Time of day changes what you see

In the same experiment, locomotion was more frequent at night. Swimming increased at night, while walking was more prominent during the day. Activity also differed among males, non-ovigerous females and ovigerous females.¹ A daytime glance and a lights-off observation therefore sample different parts of the activity cycle.

This does not mean every shrimp should follow one clock, that daytime stillness is sickness, or that artificial light duration has been optimized for home colonies. Compare at consistent times before calling a visibility change abnormal.

A separate 2024 hepatopancreas RNA series also compared 06:00, 12:00, 18:00 and 24:00, but sampling time changed together with light state, temperature and time without feed.⁷ Its unequal ten sequencing libraries and one significant qRT-PCR target can support a time-associated molecular observation under that combined history. They do not prove an internal circadian clock, explain a swimming event, predict a molt or define when lights or food should be provided.

Different shrimp can have different baselines

A 2024 study repeatedly tested 48 captive red cherry shrimp in open-field and food-shelter assays. Seven of eight measured traits showed supported among-individual differences. Conditional repeatability estimates ranged from 0.168 to 0.626, with a median of 0.353; food latency was the exception.²

The paper used the name Neocaridina heteropoda, which is preserved in its citation. The experiment does not turn a shrimp into a fixed personality label or predict health and survival. It shows why the boldest animal in a colony cannot define the species-wide response.

Waterborne context can change movement

A 2026 experiment compared control water with water previously occupied by green swordtails. No fish was present in the five-minute assay arena. The repeated-measures dataset included 80 shrimp tested six times; a parallel genetic dataset included 1,111 shrimp tested once.³

The 1,191 tested offspring represented 75 full-sibling families nested within 37 paternal half-sibling groups, all descended from one pet-trade-source colony. Full siblings also shared one rearing tank. The repeated animals always began with control water and alternated treatments, arena water was replaced after ten trials, and ages ranged from 41 to 215 days.

Predator-cue water shifted the multivariate response in the direction the authors interpreted as shyer. The mean shift across individual traits was modest at 0.237 standard-deviation units. The work did not measure capture, injury, survival, feeding, hatch, recruitment or aquarium coexistence. It shows that a cue can change behaviour without contact, not that swordtails or any other fish are safe or unsafe.

The pedigree model attributed an estimated 24.3 percent of multivariate behavioural variance to additive genetic effects, but uncertainty was high and family remained partly confounded with its rearing tank. That population estimate is not a fixed personality score, a commercial-line heritability or a prediction for one shrimp.

Does rapid swimming mean mating?

It can be compatible with search behaviour, but movement alone does not prove a freshly molted female, a pheromone or successful mating. In a 2018 experiment, juvenile females housed with adult males, adult females or alone were compared under full-contact and separated sensory conditions. Full interaction with males accelerated later ovarian growth, while visual or waterborne chemical cues at a distance did not reproduce that result.⁴

The authors proposed tactile or possible contact-associated chemical cues during male mounting and described the species as using a pure-search mating system. They did not test whether a home-aquarium swimming event was caused by a newly molted female. Confirm reproductive progress from anatomy and later outcomes: visible ovarian development, observed contact, attached eggs or a dated hatch, each with its own evidence limit. Use the sexing and life-stage guide.

Noise and vibration can change a short feeding trial

A 2023 study compared ambient playback with louder broadband playback in one test aquarium. Two experiments each used 70 mature shrimp, with 35 individual trials per playback condition. Under the louder playback, shrimp shifted farther from the speaker and showed longer food-finding latency, fewer food revisits and more feeding distraction. Total movement speed did not differ between playback conditions.⁵

The aquarium and speaker arrangement were not independently replicated, particle motion was not measured and opening a divider coincided with playback onset for part of the design. This supports recording a new pump, speaker, construction or vibration event when behaviour changes. It does not identify a safe decibel limit, prove that an ordinary filter harms shrimp or let a phone sound reading reproduce underwater exposure.

A mostly negative movement result is not a safety result

A 2021 study tested purchased White Pearl N. palmata with concentrated, particle-filtered extracts from one recycled LDPE material and one starch-blend foil. Four material-by-weathering experiments each used eight one-shrimp vessels at five concentrations and repeated movement observations on days 1, 3, 7 and 14.⁶

The extracts inhibited bacterial luminescence, but moved distance and frozen-event counts in shrimp showed high individual variation and no consistent concentration-response. This does not prove the extracts were harmless. Duplicate leachates were pooled, the treatments were source-plastic equivalents rather than identified chemical concentrations, and only two locomotor endpoints were tested. A behaviour assay can miss another biological effect, while an isolated movement difference can still fail to identify its cause.

What common observations establish

When a change needs investigation

A sudden synchronized change across several animals carries more operational weight than one individual's familiar pattern. It still has a differential, not a diagnosis. Use this sequence:

1. Confirm the observation from more than one angle without prolonged handling.
2. Check power, filter output, air delivery, heater state and room conditions.
3. Verify temperature with a second instrument if a thermal event is possible.
4. Test total ammonia nitrogen, nitrite and pH; measure dissolved oxygen when available.
5. Review source-water treatment and every recent food, plant, chemical, aerosol, water change, top-off and equipment event.
6. Restore a measured flow or aeration failure with verified clean equipment and air.
7. Stop a continuing known exposure and use the chemical-safety guide.
8. Prepare verified replacement water before an exchange; do not choose a universal percentage from behaviour alone.
9. Close sales, transfers and shared wet equipment while mortality, abnormal behaviour or cause remains unresolved.

Do not add food, minerals, salt, medication, conditioner, adsorbent or another chemical as a reflex response to movement alone. Use the filtration and oxygen guide for equipment, dissolved-oxygen and outage evidence, and the health triage guide for lesions, deaths, balance loss and diagnostic limits.

How to ask the community for useful help

• Show one wide tank view with distribution and equipment.
• Show one close view with movement quality and balance.
• State exact time, light state, duration and number affected.
• Give the feeding, maintenance, transfer, molt and tankmate timeline.
• Report test method and result, not only "parameters are perfect."
• List recent wet and airborne additions.

A defensible answer should say what was observed, what it is compatible with, which sign makes it urgent and what remains unresolved. "Looks normal" is not a useful conclusion unless the comparison and stop conditions are named.

What the evidence does not prove

• That hiding means stress or visibility means happiness
• That rapid swimming proves a female molt or sex pheromone
• That pellet crowding proves hunger
• That stillness reveals sleep, premolt, old age or disease
• That waterline clustering identifies one water problem
• That one individual's behaviour represents the colony or species
• That a familiar behaviour cannot also occur during an emergency

Primary sources

1. Carvalho-Batista et al., 2023. Shelter preference and variation in the daily activity pattern of the ornamental shrimp Neocaridina davidi. Nauplius 31: e2023018.
2. Rickward, Santostefano and Wilson, 2024. Among-individual behavioural variation in the ornamental red cherry shrimp, Neocaridina heteropoda. Ecology and Evolution 14(2): e11049.
3. Wilson, Rickward and Santostefano, 2026. Quantitative genetics of shy-bold behaviour and plastic response to novel predator cues in the cherry shrimp, Neocaridina davidi. Journal of Evolutionary Biology, voag042.
4. Tropea, Lavarías and López Greco, 2018. Getting ready for mating: The importance of male touching as an accelerator of ovarian growth in a caridean shrimp. Zoology 130: 57-66.
5. Azarm-Karnagh, López Greco and Shafiei Sabet, 2023. Annoying noise: effect of anthropogenic underwater noise on the movement and feeding performance in the red cherry shrimp, Neocaridina davidi. Frontiers in Ecology and Evolution 11: 1091314.
6. Klein, Hess, Schulte-Oehlmann and Oehlmann, 2021. Locomotor behavior of Neocaridina palmata: a study with leachates from UV-weathered microplastics. PeerJ 9: e12442.
7. Zhang et al., 2024. Transcriptome profiles of the hepatopancreas of the Chinese swamp shrimp Neocaridina denticulata under different diurnal rhythms. ScienceAsia 50(2): 2024043. Official article and supplement PDF. NCBI BioProject PRJNA832034.

Separate behaviour from tankmate survival and recruitment.

Measure feeding response without turning one pellet into a diet test.