Subjective remembering and mis-remebering: The rise of memory control in children with typical develpment and with disabilities

Remembering one’s own past is an extremely important function as it is related to the human ability of constructing an identity and a sense of Self. Despite the powerful strengths of our memory, research in the past decades has repeatedly revealed its fallacy, both in adults (e.g., Gallo, 2006) and children (e.g., Brainerd, Reyna, & Ceci, 2008). The case of children is of particular interest, given that initial studies on false memories – remembering episodes that never happened or that did happen but are now recollected in a distorted manner – have supported evidence in favour of children as being more susceptible to memory errors than older children and adults; however, subsequent investigations have revealed a developmental reversal of this trend, with children producing less false memories than adults (Brainerd et al., 2008). The general scope of this dissertation is to address the conditions under which children may evince memory distortions less or more easily than adults, and whether they are able to introspect on their memory states and differentiate true memories from false memories at the subjective level. For this purpose, I employed two techniques in order to investigate subjective remembering: The Remember-Know paradigm (Tulving, 1985) and confidence ratings about one’s own memory traces in different recognition memory paradigms. I was particularly interested in studying these phenomena both in typically developing children (Experiment 1 and Experiment 2) and in children with developmental disabilities (Experiment 3, Experiment 4 and Experiment 5). In Experiment 1 and Experiment 2, the effects of warning (i.e., telling participants what a false memory is) on false-memory rates and subjective experiences (Remember-Know judgments) were investigated in children (9- and 11-year-olds, Exp.1; 7 to 13 year-olds, Exp. 2) and young adults with the Deese-Roediger-McDermott (DRM) paradigm. This paradigm involves the presentation of lists of semantically associated words (e.g., nurse, sick, lawyer, medicine) converging in meaning on a word, the critical lure, which is not presented in the study list but represents the semantic gist of the list (i.e., doctor). After studying these word lists, adults falsely recall or recognize the critical lures with high frequency and claim to experience vivid recollection for the critical lures almost at the same rate as they do for studied words. In Experiment 1, I found that providing a warning prior to list encoding increased false recognition in both groups of children. This effect was modulated by recall: 9-year-olds produced more false recognition when they performed a recall task before the recognition task, whereas the opposite was true for 11-year-olds. The subjective experience of false memories, as assessed with the Remember-Know procedure, was similar across age groups (including young adults). In Experiment 2, two warning conditions (warning with an example of a critical lure, and warning without an example of a critical lure as in Experiment 1) were compared to a control condition. It has been found that the 7-year-olds produced more false memories in both warning conditions compared to the control condition; in contrast, 12-year-olds and adults exhibited reduced false-recognition rates in the warning conditions. The subjective experience associated with false memories was again similar across ages, but differed for true memories, such as age-related increase in subjective recollection was found. Experiment 3 addressed true memories and memory errors when material organized in scripts is presented at encoding, in a special population of children. A paradigm for scripted material (adapted from Lyons, Ghetti, & Cornoldi, 2010) and confidence ratings have been employed. Children with Attention Deficit/Hyperactivity Disorder (ADHD) and a control group of children matched for age and educational level viewed 18 photographs for each of 4 scripts (waking up in the morning, going grocery shopping, eating at a restaurant and attending a lecture at school). A recognition test followed which included old photographs and new photographs some of which presenting information consistent with the script and some of which depicting the cause (e.g., wiping the table) of an effect (e.g., knocking over a glass of coke) actually viewed during encoding. Children with ADHD, compared to control participants, exhibited lower false recognition for photographs consistent with the script (naming, gap-filling errors), but higher false recognition for causal inferences (naming, backward causal inference errors). Furthermore, children with ADHD were generally more confident in their errors than were control participants. It seems that semantic elaboration, recollective processes, and impulsivity all influence false memories’ formation in children with ADHD at both objective and subjective level. Experiment 4 and Experiment 5 were aimed at investigating recognition memory and recollective experience for a text in adolescents with and without learning difficulties. In Experiment 4, adolescents (age 15 to 19) with learning difficulties were selected upon their performance on a standardized test for text comprehension and on the teachers’ evaluations of their school achievement. In a recognition memory paradigm for text, “poor learners”, compared to a control group, were less efficient at deciding whether target sentences appeared in a previously heard narrative, thus producing fewer hits and more false alarms. Further, “poor learners” were less likely to associate Remember judgments to the target sentences, whereas both groups associated a similar level of Familiar responses to the old items. In Experiment 5 (which tested memory for individual words), poor learners performed similarly to the control group, exhibiting the same degree of recollective experiences following a deep- compared to shallow-encoding condition, and the same pattern of associations between Remember responses and memory for specific details (i.e., item color and spatial position). Together, these results show that students with learning difficulties have a less subjectively compelling memory experience related to a complex text. Overall these experiments showed that younger children (Exp. 1 and 2) and children with developmental disabilities (ADHD, Exp. 3; learning difficulties, Exp. 4) do manifest a different ability at differentiating between true and false memories at the subjective level. Future research should investigate whether this extends to different paradigms and in different populations of children with disabilities, given the important implications this introspective and metacognitive ability has when these children are required to testify in forensic contexts.

Scopo generale delle ricerche presentate in questa tesi di dottorato è quello di studiare la percezione soggettiva del ricordo episodico in bambini a sviluppo tipico e adulti (Esperimento 1 e 2) e bambini con disabilità (ADHD, Esperimento 3; difficoltà di apprendimento, Esperimento 4 e 5). Negli Esperimenti 1 e 2, si è voluto approfondire l’effetto del warning (o avvertimento) sulla creazione dei falsi ricordi, attraverso l’utilizzo del paradigma DRM (Deese-Roediger-McDermott) e l’esperienza soggettiva legata sia ai ricordi veritieri che falsi attraverso l’uso del paradigma Remember-Know (Tulving, 1985). Bambini di 9 e 11 anni (Esperimento 1), bambini dai 7 ai 13 anni (Esperimento 2) e giovani adulti hanno partecipato alla ricerca. Il paradigma DRM consiste nella presentazione di liste di parole semanticamente legate tra loro (ad es., infermiere, medicina, ospedale). Le parole di ogni lista sono inoltre legate ad un’esca critica (nell’esempio, dottore) che non viene presentata ai partecipanti. In letteratura si evince che gli adulti in un successivo compito di riconoscimento, ricordano anche la parola critica non presentata, commettendo così un falso ricordo. Avvertire i partecipanti circa la possibilità di commettere tale falso ricordo riduce questo effetto negli adulti. Nell’Esp. 1 si è trovato che i bambini commettono invece più falsi ricordi quando avvertiti. L’esperienza soggettiva (giudizi Remember-Know) non mostra invece effetti diversi in base all’età. Nel secondo esperimento, l’aggiunta di una condizione warning con un esempio specifico di che cosa sia un falso ricordo attraverso l’utilizzo di questo paradigma aumenta ancora di più i falsi ricordi nel gruppo di bambini di 7 anni, mentre li diminuisce nei ragazzini più grandi e nei giovani adulti. Inoltre, i bambini di 7 anni associano un minor numero di esperienze Remember (paradigma Remember-Know) ai ricordi veritieri rispetto agli altri bambini e agli adulti. Nell’Esperimento 3, un gruppo di bambini con ADHD è stato confrontato con un gruppo di controllo su di un paradigma di memoria con materiale organizzato in script (Lyons et al., 2010). Dopo aver codificato una serie di fotografie rappresentanti le azioni tipiche di 4 diversi script (cena al ristorante, routine del mattino, spesa al supermercato e lezione in classe), i bambini erano sottoposti ad un compito di riconoscimento in cui alcune delle foto presentate inizialmente venivano mescolate ad altre nuove. Inoltre, nella fase di codifica venivano presentate delle immagini di effetti le cui cause non venivano invece mostrate (venivano presentate solo al riconoscimento, come distrattori, il riconoscimento delle quali costituisce pertanto un falso ricordo, chiamato Inferential causal error). I bambini con ADHD producevano più errori inferenziali causali e meno errori di riconoscimento di fotografie nuove ma coerenti con gli script. Inoltre erano più sicuri quando commettevano errori di memoria rispetto ai compagni, dimostrando di essere meno in grado di fare un’introspezione sui propri stati interni di memoria. Infine, gli Esperimenti 4 e 5 erano volti ad indagare l’esperienza soggettiva del ricordo legata al riconoscimento di frasi contenute in un testo. Un gruppo di adolescenti con difficoltà di apprendimento (15-19 anni) è stato confrontato con un gruppo di controllo. I ragazzi con difficoltà producevano meno hit (riconoscimenti corretti delle frasi contenute nel testo) e più falsi allarmi. Inoltre, dal punto di vista soggettivo, i ragazzi con difficoltà di apprendimento associavano meno giudizi Remember ai riconoscimenti corretti rispetto al gruppo di controllo (Esperimento 4). L’Esperimento 5 ha testato la memoria di riconoscimento per parole isolate negli stessi gruppi di soggetti. In questo esperimento i ragazzi con difficoltà hanno dimostrato di avere una performance simile a quella del gruppo di controllo sia per quanto riguarda i riconoscimenti corretti sia per quanto riguarda i giudizi Remember-Know. Pertanto, si conclude che i ragazzi con difficoltà hanno una ridotta recollection per materiali complessi di apprendimento come ad esempio un brano. In generale, questi esperimenti mostrano come i bambini più piccoli rispetto a quelli più grandi (Esp. 1 e 2) e i bambini con disabilità rispetto ai bambini a sviluppo tipico (Esp 3 e 4) abbiano una ridotta capacità di valutare soggettivamente i propri ricordi e di differenziare i ricordi veritieri da quelli falsi.