This reader reconstructs a specific problem in music cognition: what repeated exposure actually does to a musical object in a listener. The literature does not support a simple ladder in which repetition automatically produces recognition, familiarity, learning and liking. Instead, these processes are partially dissociable. Experimental work on mere exposure shows that even very limited prior contact can alter preference for unfamiliar melodies, while recognition follows a different temporal and cognitive trajectory. Artificial musical systems further separate memory for particular exemplars from acquisition of a musical grammar: repeated presentation of a small set can strengthen recognition and preference without producing generalisation, whereas greater variability can facilitate statistical learning without equivalent preference change. Neuroimaging studies add a further distinction between neural activity associated with subjective liking and activity associated with exposure frequency. Taken together, the corpus supports a relational model in which recurrence changes the availability of musical objects but does not determine a single affective outcome. Repetition is better understood as a family of operations that can produce perceptual fluency, exemplar memory, expectation, recognition and preference in different combinations. READER The classical mere-exposure proposition becomes unusually precise when tested with music because melodies can be simultaneously perceptual objects, structured sequences and affective objects. Peretz, Gaudreau and Bonnel (1998) conducted three experiments separating liking from recognition after exposure to familiar and unfamiliar melodies. Their central result is not merely that repetition can increase liking. It is that the memory effects measured through affective judgement and explicit recognition respond differently to familiarity, delay, timbral change and task manipulation. A single repetition was sufficient to increase liking for novel melodies, while recognition was strongest for already familiar melodies; the effects also displayed different persistence across delays. The authors therefore interpret affective and recognition effects as reflecting partly distinct implicit and explicit memory processes. This is foundational for any account of repeated listening because it blocks a common shortcut: recognising a piece and preferring it are not two measurements of the same underlying state. Loui and Wessel (2008) make the separation still sharper by removing pre-existing cultural familiarity. Their Bohlen–Pierce artificial musical system exposed participants to a grammar radically different from conventional Western tonal organisation. In Experiment 1, fifteen melodies were presented twenty-seven times each. Participants subsequently recognised old melodies and generalised to new grammatical melodies, indicating acquisition of statistical structure, but preference did not differentiate familiar, novel grammatical and ungrammatical material. In Experiment 2, ten melodies were instead presented forty times each. Recognition became strong, generalisation failed, and previously heard melodies were preferred. The contrast is theoretically productive: more varied exemplars supported grammar learning, whereas more concentrated repetition favoured exemplar memory and preference. The authors explicitly resist defining a universal numerical threshold for preference change, noting that repetition interacts with duration, prior familiarity and properties of the input. Their experiments therefore support a topology rather than a staircase: repetition density and corpus diversity redistribute learning between general rules and particular objects. Loui’s subsequent work on melody and harmony reinforces the dissociation. Manipulations capable of disrupting recognition and grammatical learning could leave a mere-exposure preference effect intact, while other manipulations permitted learning without an equivalent preference effect. Preference formation can therefore occur without successful conscious recognition, and successful structural learning need not generate preference. The distinction matters beyond experimental artificial grammars. Everyday listening simultaneously exposes listeners to particular songs and to larger stylistic distributions. Repetition of one recording can stabilise an exemplar while broad exposure to a genre can train expectations about what counts as a plausible continuation, timbre, cadence, rhythm or form. These two histories overlap but remain analytically distinct. Green and colleagues (2012) provide a graded exposure design and a neural complement. Melodies were encountered at multiple frequencies—0, 1, 2, 8 and 32 previous presentations—and recognition ratings increased strongly with exposure. Exposure frequency was associated with increased activity in dorsolateral prefrontal and parietal regions, while subjective liking was associated with a different pattern including anterior insula and dorsal striatum. The authors describe supraliminal mere exposure as one contributor to positive judgement but caution that their imaging analysis requires further empirical substantiation. This separation is important: the neural signature of “having encountered this repeatedly” is not reducible to the signature of “I like this”. Salakka and colleagues (2021) broaden the object again by asking what makes real-world music memorable and emotionally effective. Their analysis of 140 songs relates acoustic musical features to familiarity, emotions and memories in older adults. Familiarity is neither an incidental nuisance variable nor a complete explanation. Musical features, emotion, familiarity and autobiographical salience interact, and the authors explicitly consider applications in which music might be selected for emotional power, familiarity and memory relevance. This moves the problem from controlled recurrence toward ecological listening: repeated exposure takes place through culturally distributed musical objects whose acoustic features and prior histories are already heterogeneous. The combined literature therefore warrants a more careful vocabulary. Exposure is an event of contact. Repetition is recurrence of an object or class of objects. Recognition is an explicit or implicit discrimination of prior encounter. Familiarity is a graded experiential quality produced by prior contact but shaped by prior knowledge and context. Statistical learning is sensitivity to regularities across exemplars. Preference is an affective evaluation. Liking can change through repetition without corresponding grammatical learning; grammar can be learned without increased liking; recognition can persist when affective effects weaken or vice versa. None of these findings licenses fixed everyday thresholds such as “ten listens produces liking” or “one hundred listens produces fixation”. What they support is a differential accumulation model in which the distribution of encounters changes what the listener can predict, recognise and value. For Socioplastics, the importance of this literature lies in the transformation of recurrence from quantity into structure. A recurrent object gains a different operational position within an environment: it becomes easier to retrieve, anticipate and situate. But recurrence also acts at the level of the surrounding grammar. A listener exposed to many heterogeneous instances can acquire a field of expectations that changes how later objects are parsed even when those particular objects are new. Musical familiarity is consequently both local and environmental: familiarity with this track and familiarity with the language within which this track becomes intelligible. The reader therefore establishes the first half of a larger hypothesis. Repetition can produce availability, fluency, recognition, expectation and preference, but none of these alone explains why a very small subset of music becomes attached to a particular period, place, relationship or state of self. That transition requires autobiographical memory and personal familiarity, developed in Reader 02. Read together, these studies replace the idea of a universal repetition threshold with a differential model of musical recurrence. Exposure changes what can be recognised, anticipated and valued, but it does not predetermine autobiographical attachment. This reader therefore stops at the boundary between familiarity and personal history. The next problem is not why a track becomes known or liked, but why certain familiar objects acquire a specific temporal, spatial and affective address.
BIBLIOGRAPHY
Anglada-Tort, M., Masters, N., Steffens, J., North, A. and Müllensiefen, D. (2023) ‘The Behavioural Economics of Music: Systematic review and future directions’, Quarterly Journal of Experimental Psychology, 76(5), pp. 1177–1194. doi:10.1177/17470218221113761.
Green, A.C., Bærentsen, K.B., Stødkilde-Jørgensen, H., Wallentin, M., Roepstorff, A. and Vuust, P. (2012) ‘Listen, Learn, Like! Dorsolateral Prefrontal Cortex Involved in the Mere Exposure Effect in Music’, Neurology Research International, 2012, 846270.
Loui, P. and Wessel, D. (2008) ‘Learning and liking an artificial musical system: Effects of set size and repeated exposure’, Musicae Scientiae, 12(2), pp. 207–230.
Loui, P. (2012) ‘Learning and liking of melody and harmony: further studies in artificial grammar learning’, Topics in Cognitive Science.
Peretz, I., Gaudreau, D. and Bonnel, A.-M. (1998) ‘Exposure effects on music preference and recognition’, Memory & Cognition, 26(5), pp. 884–902.
