How newborn animals use weak preferences to make early decisions

Within hours of being born or hatching, newborn animals often face vital decisions. Chicks follow their mothers, while turtle hatchlings head towards safer places almost straight away, well before experience or learning can influence their actions.

New research suggests these first choices are shaped neither by inflexible instincts nor by empty minds. Instead, young animals draw on a subtle set of innate, adaptable preferences that work together to direct behaviour.

A study led by researchers at Queen Mary University of London shows how the combination of numerous weak signals helps newborns make accurate choices in an uncertain world.

Newborn animals begin acting almost at once

Within hours of birth or hatching, some animals start to move about and explore. Scientists describe these animals as precocial.

Chicks, ducklings, turtle hatchlings and many grazing mammals belong to this group. Because threats arise early, their survival relies on making quick, accurate decisions.

Earlier scientific theories proposed that newborn minds begin empty. Evidence gathered from many species now presents a different view.

Newborn animals come into the world with several innate preferences. Although each one is weak and adaptable, together they help guide useful actions.

Soft preferences shape decisions

Traditional studies of animal behaviour concentrated on powerful instincts, including automatic responses to escape or mate.

Such instincts are helpful in emergencies, but they can lead to errors when circumstances change. The new research draws attention to a different kind of bias, known as early predispositions.

These predispositions take the form of gentle preferences rather than rigid rules. A chick, for instance, might be mildly drawn to particular colours, show some interest in upward movement or respond slightly to sound.

No individual signal compels a response. Preferences vary between individuals and frequently disappear after early life.

That flexibility has a benefit. Weak signals help limit false alarms: rather than responding to a single deceptive cue, newborn animals wait for several signs to match.

Uncommon features limit confusion

The research finds that many early preferences centre on features seldom seen in natural surroundings. Red and blue colours, for example, are uncommon in soil and plants.

Movement upwards against gravity generally indicates a living creature. Biological motion also distinguishes itself from the movement of objects.

Unusual features can be highly meaningful. Being drawn to rare cues reduces the likelihood of confusion because a random object is unlikely to display several such features together. A parent animal, by contrast, often displays many of them.

Bringing small clues together

Combination is the central idea in the paper. Young animals bring multiple weak preferences together at the same time, drawing information from colour, size, movement, shape and sound.

A mother hen, for instance, often has reddish colouring and a large size, moves upwards and changes speed. On its own, each signal is uncertain.

Together, however, the signals provide strong evidence. When clues occur at the same time across different senses, coincidence becomes less likely.

The researchers call this a self-supervised strategy. It involves neither teaching nor feedback: the brain instead draws on the structure of the environment itself.

How signals strengthen one another

To explain how cues come together, the paper presents a mathematical model. One method simply adds the signals.

Under this approach, each cue makes a small contribution to a decision. Additive processing is most effective in straightforward surroundings with few distractions.

A second method relies on synergy, which occurs when cues amplify one another. Signals from different senses, such as sight and sound, increase confidence far more than signals from a single sense alone.

Independent cues are unlikely to line up by chance. Synergy therefore helps preserve accuracy, even in noisy surroundings.

Experiments indicate that synergy has an important part to play in early life. Chicks are more strongly attracted when colour and movement are presented together, while pairing sound with motion improves accuracy further.

Preventing false alarms early in life

At first, weak biases can seem inefficient, while strong instincts appear safer. But strong instincts can trigger a response to even a partial match; weak biases require confirmation.

This system shields young animals from costly mistakes. Their early surroundings contain many misleading objects, and waiting for several matching cues helps them avoid danger and unnecessary energy use.

Flexibility also leaves room for learning later on. Early predispositions direct attention towards useful targets, after which learning mechanisms such as imprinting reinforce bonds with parents or food sources.

From newborn animals to machines

Animal research also informs human science. Human infants display comparable early preferences for faces and biological motion. Understanding how decisions begin before learning takes place benefits developmental psychology.

Artificial intelligence faces its own difficulties when training data is scarce. Modern systems often require enormous numbers of examples.

Biological strategies point to an alternative: machines could combine cues and use weak built-in preferences to act with less training.

Subtle choices support newborn survival

“Reviewing the literature on inexperienced animals helped to identify an underlying reason for why animals exhibit ‘soft’ preferences,” said study co-author Elisabetta Versace, a senior lecturer in psychology at Queen Mary University of London.

Weak preferences might initially appear ineffective in early life, when strong instincts seem safer and faster to act on. The research discussed in the study, however, indicates the opposite advantage.

The authors note that weak, temporary preferences can limit false alarms in decision-making. This allows animals to assess several environmental cues instead of responding too rigidly to just one signal.

Uncertainty in early life

By modelling the phenomenon, the researchers could also reinterpret more than a century of experimental work.

Results that once seemed scattered or inconsistent now look like partial clues to how organisms deal with uncertainty in early life.

The model also makes clear predictions that can be tested in animals and machines in future.

An understanding of early intelligence shows that simple rules can give rise to surprisingly complex behaviour. Even at birth, life may already possess a quiet kind of wisdom.