Državna Matura Psychology common mistakes
Državna Matura Psychology · Understand the traps and how to avoid them.
When asked to identify the research method used in a described study, the candidate labels it 'eksperiment' whenever the words 'istraživanje' or 'ispitivanje' appear, without checking whether the researcher actively manipulated an independent variable and randomly assigned participants to conditions.
The word 'experiment' is used loosely in everyday Croatian, and the candidate pattern-matches on the topic (e.g. memory, stress) rather than on the design logic. The manipulation-and-random-assignment criteria are learned as a definition but not applied as a checklist while reading the scenario.
What to do instead
Before choosing, extract three facts from the scenario: (1) Did the researcher change something deliberately? (2) Were participants randomly allocated to levels of that variable? (3) Was there a control condition? If any answer is no, it is korelacijsko or deskriptivno research, not an experiment. Practise by rewriting five short scenarios into these three slots before answering.
The candidate reads a correlation coefficient and concludes that one variable causes the other, or treats a correlation of r = 0.30 as proof of no relationship at all.
Correlation and causation are conflated because the numerical direction feels like an explanation. The second error comes from treating any r below about 0.5 as 'nothing', which misreads the meaning of effect size and ignores sample size and measurement reliability.
What to do instead
Apply two fixed sentences: 'Correlation tells us about co-variation and prediction direction, not about mechanism.' And judge magnitude by the squared value plus context, not by a single threshold. Also name the third-variable possibility explicitly in any answer that discusses correlation, since examiners award marks for identifying plausible confounding factors.
The candidate describes the structure and function of a brain region using only a memorised label ('the hypothalamus controls emotions') and cannot say what happens when it is damaged or which system it belongs to.
Brain anatomy is studied as a list of names paired with one keyword, so retrieval produces the keyword but not a functional model. The candidate also confuses nearby structures (hypothalamus versus thalamus, amygdala versus hippocampus) because they were learned in the same study session without contrast.
What to do instead
For each structure, prepare a three-part card: location and system, core function in one sentence, and a concrete consequence of damage or stimulation. Study confusable pairs side by side (thalamus relays sensory input; hypothalamus regulates internal state via the endocrine and autonomic systems; amygdala attaches emotional significance; hippocampus consolidates declarative memory).
The candidate answers questions about sensation and perception by treating them as synonyms, so a question about sensory transduction is answered with a description of perceptual organisation, or the reverse.
The two terms are colloquial synonyms in Croatian usage, and textbook chapters present them consecutively, so the candidate's mental boundary between receptor-level detection and interpretation is blurred.
What to do instead
Anchor a one-line distinction: sensation is the detection and encoding of physical energy by receptors; perception is the selection, organisation and interpretation of that information. Then classify every example you meet by asking 'could this happen with a single receptor firing, or does it require combining and interpreting input?'
When a question presents a learning scenario, the candidate names a conditioning concept from memory without checking the temporal order of stimuli, so classical and operant conditioning are mixed up and terms like 'reinforcement', 'punishment' and 'extinction' are applied to the wrong events.
The vocabulary is learned as a set of labels rather than as a causal structure. Under time pressure the candidate grabs the most familiar term, and negative reinforcement is especially often misread as punishment because of the word 'negative'.
What to do instead
For every scenario, write out the sequence: what comes before the behaviour, what the behaviour is, and what follows it. If the key event precedes the behaviour and is involuntary, it is classical conditioning. If a consequence changes the frequency of a voluntary behaviour, it is operant. Remember negative reinforcement removes an aversive stimulus and increases behaviour; punishment decreases it.
The candidate answers questions on memory by treating short-term memory, working memory and long-term memory as one store, and cannot distinguish encoding, storage and retrieval failures when explaining forgetting.
The three-stage model is memorised as a diagram but not as a set of separable processes, so decay, interference and retrieval failure all get labelled 'forgetting' without further specification.
What to do instead
Use a two-axis frame: which store (sensory, short-term/working, long-term) and which process (encoding, storage, retrieval). Then map each classic explanation onto it: decay is a storage problem over time, interference is a retrieval or consolidation problem from competing material, and cue-dependent forgetting is a retrieval problem. Answer with both axes named.
The candidate equates intelligence with a single IQ number and answers questions about theories of intelligence by naming one theorist, without being able to contrast psychometric, multiple-intelligences and developmental approaches.
Popular culture and school testing reinforce the single-score image, and the exam syllabus contains several competing models that were studied as separate lists rather than as a debate about what intelligence is.
What to do instead
Prepare a comparison table with columns for the core claim, the evidence used, and the main criticism for each approach (general-factor psychometrics, multiple independent abilities, multiple intelligences, triarchic or developmental views). For any question, first decide whether it asks about structure, about prediction, or about development, then choose the model that addresses that question.
The candidate labels every strong emotional reaction as 'stress' or every low mood as 'depression', and answers clinical vignettes by matching a single symptom to a diagnosis instead of checking duration, impairment and the required symptom cluster.
Everyday language uses clinical terms loosely, and vignettes are read for the most vivid symptom. The candidate skips the diagnostic thresholds because they feel like detail compared with the narrative.
What to do instead
For every vignette, note four things before choosing: duration, number and type of required symptoms, functional impairment, and exclusion of other causes such as substances or medical conditions. Treat a single symptom as insufficient by definition, and be ready to say why a diagnosis is not justified when criteria are unmet.
The candidate answers developmental questions by describing what children of a given age typically do, without linking the behaviour to the stage or mechanism the question is testing, and mixes up Piagetian stages, ages and key achievements.
Ages and stage names are memorised in isolation, and the candidate relies on general impressions of what children can do rather than on the specific construct (object permanence, conservation, egocentrism, theory of mind) that the item targets.
What to do instead
Build a stage table with columns: stage, approximate age range, key achievement, classic task, and what the child cannot yet do. When answering, identify which construct the question is about first, then recall the stage that construct belongs to, rather than starting from the age.
The candidate explains social behaviour using dispositional terms only ('he is aggressive', 'she is conformist') and ignores situational and group-level explanations such as norms, roles, obedience pressure or attribution biases.
The fundamental attribution error is not only a topic in the syllabus but also a habit of everyday reasoning, so the candidate reproduces the bias while answering questions about it. Social psychology items often hinge on recognising situational pressure.
What to do instead
For any social scenario, force yourself to generate one dispositional and one situational explanation before selecting an answer, and check whether the question names a specific phenomenon (conformity, obedience, bystander effect, group polarisation, attribution error). If the item describes strong situational cues, the situational account is usually the intended one.