EVOPEDIA

← all systems · see it in The System

Fish school (shoal)

In biology, any group of fish that stay together for social reasons are shoaling, and if the group is swimming in the same direction in a coordinated manner, they are schooling.

Properties published scienceour proposal

Sourced fields, and our own readings.

aliasesSchooling fish (members of a school)
born atkind: at least the Eocene (fossil shoal of Erismatopterus levatus); earlier not dated in the sources read. One school: formation in minutes at dusk (Georges Bank herring)
born precisionepoch-level (kind); hours (one school)
boundaryInside: the member fish while they stay together by social attraction (the shoal), and, when they also move in one direction in coordination, the school; their interaction rules (repulsion, alignment, attraction), the interaction network those rules make, the group's structure (spacing, density, polarity, shape) and its collective states and decisions. Outside: the water and its light, flows and sounds; the members' own bodies and brains as organisms (walked as sys:fish); predators, prey, cleaners and fishers; the fish stock or population the school belongs to.
nested levelA collective of fish: part of a fish stock or population (schools of a stock travel between spawning, feeding and nursery grounds), and of the pelagic food web it feeds. Shoaling is a special case of aggregating, and schooling a special case of shoaling.
scalepopulation

Measures (28) published science

Share of fish species that shoalabout one quarter all their lives; about one half part of their lives fraction of species
Energy saving by position8.8 (ahead of neighbour) to 19.4 (behind) % less than swimming alone
Energy difference behind vs aheadabout 10.6 % less energy per unit time at the most posterior position
Robot power spread with phaseabout 13.4 % between maximum saving and maximum cost
Share of possible wake saving captured15 % of the total possible (goldfish pairs)
Time in energy-saving phase59 % of time (goldfish followers at D = 0)
Calming effect on minimum metabolic rate25.9 (range 5-60) % reduction in a shoal
Time polarised by group size0.57, 0.26, 0.18, 0.05 for 30, 70, 150, 300 fish fraction of time
State transition rate2.0, 1.4, 1.2, 0.8 for 30, 70, 150, 300 fish transitions per minute
Longest milling bout17, 110, 708, 1245 for 30, 70, 150, 300 fish
Biological-motion tuning1.2 ± 1.6 (juvenile bout frequency about 1.25) Hz
Bait ball sizeup to 20 m diameter
Largest herring schoolup to 4.8 km3 at 0.5-1.0 fish per m3, about three billion fish (Radakov) km3; fish
Bait ball lifetimeseldom longer than 20 min
School densityabout one fish per cube of body length fish per body length cubed
Sardine run shoal7 km long, 1.5 km wide, 30 m deep (a larger shoal) km; m
Migration after the density transitionup to 40 km
Fish wave displayup to 2 min, many thousands of fish; doubled the wait to the next attack min; ratio
Schooling recovery after lateral line ablation4-8 weeks (hair cells regenerated after 1 week)
Herring burst-pulse sound1.7-22 kHz pulse trains lasting 0.6-7.6 s kHz; s
Copepod re-arming time60 ms; about 80 darts before it tires ms; count
Stickleback schooling position QTLjust over 20% of head x-position variance (two QTL) % variance
Schooling QTL effect sizes (Astyanax)6.4% (with dark preference, LG27); 12% (vision-independent, LG6) % variance explained
Tectal ablation effectmore than 95% fewer DT-BPNs (94 ± 4 to 4 ± 6 per animal) cells per animal
Shoal size imagedtens of millions of fish over many kilometres
Group sizes in lab cascadesabout 150 fish
Self-propelled particle order exponentβ ≈ 0.45 (order scales as (η_c − η)^β) dimensionless
Attraction onset in zebrafishfrom 7 days post-fertilisation, rising to 3 weeks

Open contentions (6) open contention

Where sources or readings disagree. Both stay on the map.

Partners and parts published science

Built from / builds (7)

Sits inside / holds (3)

Trades with (valence) (12)

expressed by (14)

floor filled by (12)

Kinds (9)

Notes published science

Layers

The ways it takes in its world, where they meet, and how it acts back, each from L1 (where it meets the world) up.

the boardroom (shared floors)

  1. L5: The interaction network: weighted, directed, heterogeneous links set by what each fish sees (visual networks fit best); responses spread in waves; information from a few informed members propagates
  2. L6: Collective state and structure: swarm, milling, polarised (bistable when ordered); density self-regulated (H-stable); structure encodes risk and acts as collective memory
  3. L7: Collective decision and action: quorum responses, consensus with an informed minority, emergent sensing of gradients, the school's evasion, travel, foraging and spawning

the school as collective-behaviour research describes it (reader's view: the shared water world, the members' senses, each fish's reading of its neighbours, the interaction rules, the interaction network, the collective state, collective decisions and actions)

  1. L1: The shared water world: light carrying neighbours' bodies and motion, the flows and wakes neighbours make, sound, odours (including alarm substance), light and shade gradients
  2. L2: The members' senses: eyes on the sides of the head (with schooling marks on the neighbours to read), lateral lines, nose; each member is a sensor
  3. L3: Each member's brain picks out neighbours: retina to optic tectum to dorsal thalamus neurons tuned to fish-like bout motion, insensitive to global motion
  4. L4: The interaction rules each fish runs: short-range repulsion, long-range attraction, speed regulation (alignment emerges); responses not a plain average of neighbours
  5. L5: The interaction network: who responds to whom (visual, weighted, directed, heterogeneous); behaviour spreads through it as waves and cascades
  6. L6: The collective state and structure: swarm, milling or polarised; density held constant as the group grows; structure that encodes risk and keeps a collective memory
  7. L7: Collective decisions and action back on the world: quorum and consensus choices, guidance by a few informed members, sensing of gradients no member can track, evasion, migration, foraging, spawning

way in: Alarm substance and shoal-mate odour

  1. L1: Schreckstoff: chondroitins released from torn skin near a successful predation
  2. L2: Olfactory cues from shoal-mates taken in (with sight) lower metabolic rate
  3. L4: Members close up: nearest-neighbour distance drops sharply on first exposure; startles become more frequent

way in: Light and shade over the habitat (a scalar each fish reads locally)

  1. L1: A complex environmental light gradient
  2. L2: Each fish measures only the local light level (a scalar, no direction)
  3. L4: Each fish changes speed with local light

way in: Members' own state (hunger, digestion, aerobic scope, stress)

  1. L1: A meal and its digestion; oxygen use
  2. L2: Digestion uses up aerobic capacity for swimming
  3. L4: Position taken in the school: hungry fish forward, fed fish drift back; lower aerobic scope shuffles to the rear

way in: Neighbours' near-field flows (lateral line)

  1. L1: Water currents and vibration made by nearby fish
  2. L2: Neuromasts: hair cells that respond to water movement
  3. L4: Position and spacing: without lateral lines fish swim closer; blinded saithe keep station by lateral line; lateral-line QTL overlap schooling-position QTL

way in: Neighbours' wakes (vortices taken in as energy)

  1. L1: Reverse von Karman vortices shed by fish ahead; high pressure in front of a swimmer
  2. L2: The follower's body and tail meet the induced flow
  3. L3: Tail-beat phase set linearly by front-back distance (vortex phase matching)
  4. L4: Lower tail-beat frequency and cost at every position; side by side at high speed

way in: Sight of neighbours, predators and light

  1. L1: Light carrying neighbours' bodies, schooling marks and movement; vision fails after dark
  2. L2: Lateral eyes; schooling marks on neighbours as reference points
  3. L3: Retina to optic tectum periventricular neurons to the dorsal thalamus biological-motion neurons
  4. L4: Attraction toward fish-like motion (repulsion kept by a separate path); each fish's visual neighbourhood

way in: Sound in the water (predators; herring's own burst pulses)

  1. L1: Herring burst-pulse sounds (1.7-22 kHz), mostly at night; predators' sounds
  2. L2: Herring hearing; schools react very fast to a predator

way out: Burst-pulse sounds (herring)

  1. L1: Pulse trains released with bubbles from the anal duct region
  2. L6: More sound per fish at higher density

way out: Collective evasion (startle waves, vacuole, bait ball, surface waves)

  1. L1: Water and surface: conspicuous rhythmic surface waves by thousands of fish for up to 2 min
  2. L5: The startle cascade runs through the network
  3. L6: The group tightens (bait ball) or opens round the threat (vacuole)
  4. L7: The school responds as one to attack

way out: Group foraging

  1. L1: Copepods caught by a grid of herring
  2. L6: Feeding schools break ranks into shoals; ram-feeding grids
  3. L7: Many eyes find food; feeding in one fish starts searching in others

way out: Spawning aggregation

  1. L1: They spawn at night on shallow parts of the bank
  2. L7: Shoals form vast spawning gatherings at dusk

way out: Travel and migration

  1. L1: Schools travel up to 40 km across the bank to spawn
  2. L6: The polarised state carries net movement
  3. L7: Organised group migration after the transition; small sets of leaders sway much larger groups

Who fills each layer

Sources (33) published science

Rosenthal SB, Twomey CR, Hartnett AT, Wu HS, Couzin ID (2015) Revealing the hidden networks of interaction in mobile animal groups allows prediction of complex behavioral contagion. PNAS 112:4690-4695, doi:10.1073/pnas.1420068112
Shoaling and schooling (Wikipedia, via Wikimedia Enterprise)
Kappel JM et al. 2022. Visual recognition of social signals by a tectothalamic neural circuit. Nature 608:146-152. doi:10.1038/s41586-022-04925-5 (behind its links)
Makris NC et al. 2009. Critical population density triggers rapid formation of vast oceanic fish shoals. Science 323:1734-1737. doi:10.1126/science.1169441 (behind its links)
Sosna MMG et al. 2019. Individual and collective encoding of risk in animal groups. PNAS 116:20556-20561. doi:10.1073/pnas.1905585116 (behind its links)
Li L, Nagy M, Graving JM, Bak-Coleman J, Xie G, Couzin ID. 2020. Vortex phase matching as a strategy for schooling in robots and in fish. Nat Commun 11:5408. doi:10.1038/s41467-020-19086-0 (behind its links)
McLean S, Persson A, Norin T, Killen SS. 2018. Metabolic costs of feeding predictively alter the spatial distribution of individuals in fish schools. Curr Biol 28:1144-1149. doi:10.1016/j.cub.2018.02.043 (behind its links)
Tunstrøm K, Katz Y, Ioannou CC, Huepe C, Lutz MJ, Couzin ID. 2013. Collective states, multistability and transitional behavior in schooling fish. PLoS Comput Biol 9:e1002915. doi:10.1371/journal.pcbi.1002915 (behind its links)
Berdahl A, Torney CJ, Ioannou CC, Faria JJ, Couzin ID. 2013. Emergent sensing of complex environments by mobile animal groups. Science 339:574-576. doi:10.1126/science.1225883 (behind its links)
Ioannou CC, Guttal V, Couzin ID. 2012. Predatory fish select for coordinated collective motion in virtual prey. Science 337:1212-1215. doi:10.1126/science.1218919 (behind its links)
Ward AJW, Sumpter DJT, Couzin ID, Hart PJB, Krause J. 2008. Quorum decision-making facilitates information transfer in fish shoals. PNAS 105:6948-6953. doi:10.1073/pnas.0710344105 (behind its links)
Greenwood AK, Wark AR, Yoshida K, Peichel CL. 2013. Genetic and neural modularity underlie the evolution of schooling behavior in threespine sticklebacks. Curr Biol 23:1884-1888. doi:10.1016/j.cub.2013.07.058 (behind its links)
Fish (Wikipedia, via Wikimedia Enterprise snapshot 2026-11-01) (behind its links)
Marras S et al. 2015. Fish swimming in schools save energy regardless of their spatial position. Behav Ecol Sociobiol. doi:10.1007/s00265-014-1834-4 (behind its links)
Nadler LE, Killen SS, McClure EC, Munday PL, McCormick MI. 2016. Shoaling reduces metabolic rate in a gregarious coral reef fish species. J Exp Biol 219:2802-2805. doi:10.1242/jeb.139493 (behind its links)
Doran C et al. 2022. Fish waves as emergent collective antipredator behavior. Curr Biol 32:708-714. doi:10.1016/j.cub.2021.11.068 (behind its links)
Mekdara PJ, Schwalbe MAB, Coughlin LL, Tytell ED. 2018. The effects of lateral line ablation and regeneration in schooling giant danios. J Exp Biol 221:jeb175166. doi:10.1242/jeb.175166 (behind its links)
Wilson B, Batty RS, Dill LM. 2004. Pacific and Atlantic herring produce burst pulse sounds. Proc R Soc B 271 Suppl 3:S95-S97. doi:10.1098/rsbl.2003.0107 (behind its links)
Kowalko JE et al. 2013. Loss of schooling behavior in cavefish through sight-dependent and sight-independent mechanisms. Curr Biol 23:1874-1883. doi:10.1016/j.cub.2013.07.056 (behind its links)
Makris NC et al. 2006. Fish population and behavior revealed by instantaneous continental shelf-scale imaging. Science 311:660-663. doi:10.1126/science.1121756 (behind its links)
Vicsek T, Czirók A, Ben-Jacob E, Cohen I, Shochet O. 1995. Novel type of phase transition in a system of self-driven particles. Phys Rev Lett 75:1226-1229 (arXiv:cond-mat/0611743) (behind its links)
Hinz RC, de Polavieja GG. 2017. Ontogeny of collective behavior reveals a simple attraction rule. PNAS 114:2295-2300. doi:10.1073/pnas.1616926114 (behind its links)
Katz Y, Tunstrøm K, Ioannou CC, Huepe C, Couzin ID. 2011. Inferring the structure and dynamics of interactions in schooling fish. PNAS 108:18720-18725. doi:10.1073/pnas.1107583108 (behind its links)
Couzin ID, Krause J, Franks NR, Levin SA. 2005. Effective leadership and decision-making in animal groups on the move. Nature 433:513-516. doi:10.1038/nature03236 (behind its links)
Social discounting enables fast and reliable collective escape (preprint, arXiv:2608.14965, 2026; authors incl. ZP Kilpatrick) (behind its links)
Pitcher TJ, Partridge BL, Wardle CS (1976) A blind fish can school. Science 194:963-965. doi:10.1126/science.982056 (behind its links)
Fish aggregating device (Wikipedia, via Wikimedia Enterprise snapshot 2026-11-01) (behind its links)
Ashraf I et al. 2017. Simple phalanx pattern leads to energy saving in cohesive fish schooling. PNAS 114:9599-9604. doi:10.1073/pnas.1706503114 (behind its links)
Strandburg-Peshkin A et al. 2013. Visual sensory networks and effective information transfer in animal groups. Curr Biol 23:R709-R711. doi:10.1016/j.cub.2013.07.059 (behind its links)
Couzin ID, Krause J, James R, Ruxton GD, Franks NR. 2002. Collective memory and spatial sorting in animal groups. J Theor Biol 218:1-11. doi:10.1006/jtbi.2002.3065 (behind its links)
Larsch J, Baier H. 2018. Biological motion as an innate perceptual mechanism driving social affiliation. Curr Biol 28:3523-3532. doi:10.1016/j.cub.2018.09.014 (behind its links)
Mizumoto N, Miyata S, Pratt SC. 2019. Inferring collective behaviour from a fossilized fish shoal. Proc R Soc B 286:20190891. doi:10.1098/rspb.2019.0891 (behind its links)
Patch A et al. 2022. Kinematic analysis of social interactions deconstructs the evolved loss of schooling behavior in cavefish. PLoS ONE 17:e0265894. doi:10.1371/journal.pone.0265894 (behind its links)