A-not-B Error: The Persistence of Habit

The A-not-B error (also known as the A-not-B search error or stage 4 error) is a phenomenon in infant cognitive development where infants consistently reach for a hidden object at location A, even after watching it being hidden at a new location B. This error reveals important insights about memory, inhibition, and the developing brain. This phenomenon is related to object permanence and developmental psychology.

Table of Contents

1. What Is the A-not-B Error?
2. The Classic Procedure
3. Why Does It Happen?
4. Factors Affecting the Error
5. Theoretical Explanations
6. Key Takeaways

What Is the A-not-B Error?

The A-not-B error occurs when infants who have successfully found a hidden object at location A multiple times continue to search at A even after watching the object being hidden at a new location B. This error is most common in infants aged 8-12 months.

The Paradox

The Puzzle:

• Infants clearly see the object hidden at B
• They are capable of finding hidden objects
• Yet they persistently reach to the wrong location
• The error is robust and reproducible

The Classic Procedure

Setup

Materials:

• Two identical hiding locations (A and B)
• Attractive toy or object
• Cloth or cover for hiding
• Infant seated at table

Procedure

Phase 1: Hiding at A

1. Show infant the attractive toy
2. Hide toy at location A (in full view)
3. Allow infant to search
4. Repeat 3-5 times successfully

Phase 2: The Switch to B

1. Show infant the toy again
2. Hide toy at location B (in full view)
3. Infant sees the hiding process
4. Measure where infant searches

The Error

Typical Response:

• 8-10 month olds: ~70-80% reach to A (Diamond, 1985)
• 12 month olds: ~40-50% reach to A
• 2-year-olds: Rarely make the error

Variations:

• Delay between hiding and search
• Number of A trials
• Distance between A and B
• Distractions during hiding

Why Does It Happen?

Memory Limitations

Working Memory:

• Limited capacity to hold new location
• Previous success at A dominates
• Cannot update representation
• Habit interferes with current information

Proactive Interference:

• Previous learning interferes with new learning
• Strong memory for A location
• Weak memory for B location
• Competition between memories

Inhibition Deficits

Prepotent Response:

• Reaching to A becomes automatic
• Cannot inhibit learned response
• Habit is stronger than current perception
• Impulsive responding

Executive Function:

• Prefrontal cortex immature
• Inhibition control developing
• Cannot override prepotent response
• Gradual improvement with age

Attention Factors

Cue Competition:

• Location A has strong attentional pull
• Previous success creates salience
• B location competes for attention
• Attention captured by habit

Distraction:

• Delay increases error rate
• Attention wanders during delay
• Cannot maintain focus on B
• Working memory decays

Factors Affecting the Error

Age

8-10 months:

• Peak error rate
• ~70-80% reach to A (Diamond, 1985)
• Working memory limited
• Inhibition immature

12 months:

• Error rate decreases
• ~40-50% reach to A
• Improved memory
• Better inhibition

18+ months:

• Rarely make error
• Mature object permanence
• Executive function developed
• Flexible responding

Delay

No Delay:

• Lower error rate
• Immediate search helps
• Working memory not taxed
• Attention still on B

3-5 second delay:

• Higher error rate
• Working memory taxed
• Attention shifts to A
• Habit takes over

10+ second delay:

• Very high error rate
• Memory for B fades
• A location dominates
• Infant confused

Number of A Trials

Few A Trials (1-2):

• Lower error rate
• Habit not as strong
• B location can compete
• Easier to switch

Many A Trials (5+):

• Higher error rate
• Strong habit established
• A location very salient
• Difficult to override

Context

Same Context:

• Higher error rate
• Context cues A location
• Same room, same setup
• Habit strongly cued

Changed Context:

• Lower error rate
• Context change breaks habit
• New environment
• Fresh start

Theoretical Explanations

Piaget's View

Stage 4 Limitation:

• Incomplete object permanence
• Cannot represent invisible displacements
• Limited mental representation
• Gradual construction of reality

Critique:

• Infants show surprise at violations
• Earlier competence than search suggests
• Looking vs. reaching dissociation
• Not just conceptual deficit

Dynamic Systems Theory

Multiple Factors:

• Memory, inhibition, attention, motor skills
• All contribute to performance
• Development is multidimensional
• Context-sensitive behavior

Self-Organization:

• System settles into attractor state (A)
• Needs energy to switch to B
• Previous success strengthens A attractor
• Development changes dynamics

Connectionist Models

Neural Network Approach:

• A location has stronger weights
• B location has weaker activation
• Competition between locations
• Learning strengthens correct connections

Development:

• Gradual weight changes
• Experience modifies connections
• Maturation improves processing
• Error decreases with development

Embodied Cognition

Body in Space:

• Reaching to A creates motor memory
• Body position affects search
• Egocentric coding dominates
• Action shapes cognition

Dynamic Interaction:

• Infant-environment coupling
• Previous actions constrain future actions
• Perception-action loops
• Situated cognition

Key Takeaways

• Robust Phenomenon: A-not-B error is highly reproducible
• Multiple Causes: Memory, inhibition, attention, and motor factors
• Developmental Marker: Decreases with age and cognitive development
• Not Just Conceptual: Perseverative reaching not just lack of understanding
• Context-Dependent: Error rate varies with task conditions
• Theoretical Debate: Different explanations emphasize different factors
• Window into Development: Reveals workings of infant mind