خاطرات نادرست عاطفی در کودکان با ناتوانی های یادگیری

Research has shown that children with learning disabilities (LD) are less prone to evince associative illusions of memory as a result of impairments in their ability to engage in semantic processing. However, it is unclear whether this observation is true for scripted life events, especially if they include emotional content, or across a broad spectrum of learning disabilities. The present study addressed these issues by assessing recognition memory for script-like information in children with nonverbal learning disability (NLD), children with dyslexia, and typically developing children (N = 51). Participants viewed photographs about 8 common events (e.g., family dinner), and embedded in each episode was either a negative or a neutral consequence of an unseen action. Children's memory was then tested on a yes/no recognition task that included old and new photographs. Results showed that the three groups performed similarly in recognizing target photographs, but exhibited differences in memory errors. Compared to other groups, children with NLD were more likely to falsely recognize photographs that depicted an unseen cause of an emotional seen event and associated more “Remember” responses to these errors. Children with dyslexia were equally likely to falsely recognize both unseen causes of seen photographs and photographs generally consistent with the script, whereas the other participant groups were more likely to falsely recognize unseen causes rather than script-consistent distractors. Results are interpreted in terms of mechanisms underlying false memories’ formation in different clinical populations of children with LD.

Memory illusions occur quite frequently in the laboratory as well as in real life, and their developmental trajectory depends on the nature of the illusion. When memory illusions stem from processing semantically related information (e.g., Deese–Roediger–McDermott (DRM) paradigm: Brainerd, Holliday, & Reyna, 2004; connected–meaning paradigms: Howe, 2006), typically developing children often demonstrate age-related increases in the frequency of memory errors (Brainerd et al., 2008a and Brainerd et al., 2011; but see Ghetti et al., 2002 and Lampinen et al., 2006). The DRM paradigm involves the presentation of several lists of semantically related words (e.g., sick, nurse, medicine). Within each list, all words are associated to a single word not presented during encoding, the critical lure (in this example, doctor). At retrieval, adults and older children falsely recognize the critical lures almost at the same rate as target words. Younger children recognize the critical lures to a lesser extent than older children and adults, thus committing fewer memory errors. This developmental trend persists even when emotionally charged stimuli are used instead of neutral stimuli ( Brainerd et al., 2004 and Howe, 2007; Howe, Candel, Otgaar, Malone, & Wimmer, 2010, Experiment 2). However, it is unclear to what extent these observed trends may apply to memory for life events, as well as to special clinical populations. Most of the available research on false memory formation in certain clinical populations of children primarily focuses on the implication of eyewitness memory or suggestibility in maltreated children (Howe, Cicchetti, Toth, & Cerrito, 2004), children within autism spectrum disorders (McCrory, Henry, & Happè, 2007), or other severe developmental disabilities, such as intellectual disabilities (Brown et al., 2012, Henry and Gudjonsson, 1999, Henry and Gudjonsson, 2004 and Young et al., 2003). The majority of these studies are aimed at evaluating the best way to administer a cognitive interview when children are either victims or witnesses of abuse and are required to provide allegations in the courtroom. However, little is known about the functioning of basic underlying memory processes in these populations. Furthermore, only a few studies have investigated memory accuracy and memory distortions in special populations of children with academic difficulties, such as children with reading comprehension difficulty (Mirandola et al., 2011 and Weekes et al., 2008), children with attention deficit/hyperactivity disorder (ADHD; Mirandola, Paparella, Re, Ghetti, & Cornoldi, 2012), and children with general learning disabilities (Brainerd, Forrest, Karibian, & Reyna, 2006). Research conducted with the DRM paradigm (Roediger & McDermott, 1995) has shown that children with learning disabilities produce fewer memory errors than typically developing children (Brainerd et al., 2006 and Weekes et al., 2008). This result has been associated with lower semantic processing abilities that prevent children with disabilities from relying on the gist of the wordlists, thus mimicking the pattern of errors found when comparing younger to older children. It must be noted that these studies focused specifically on memory errors, and it is unclear whether semantic processing difficulties associated with learning disabilities may influence the way in which children retain correct information as well. A paradigm specifically designed to investigate recognition memory and text-based detail recollection in adolescents with poor semantic text processing ability revealed an overall impaired recognition of sentences and related recollection (i.e., lower proportion of Remember judgments in association with correctly recognized target sentences) in these students, compared to a control group, despite both groups displaying similar recognition memory for isolated words (Mirandola et al., 2011). The present study was designed to investigate memory accuracy and errors in two distinct groups of children with learning disabilities, namely children with nonverbal learning disability (NLD) and children with dyslexia. Children with dyslexia are characterized by poor reading decoding abilities and sometimes show reduced semantic processing abilities (Betjemann & Keenan, 2008), as confirmed by recent studies combining fMRI and ERPs (Schulz et al., 2008 and Shulz et al., 2009). Whereas children with dyslexia present both language and reading skills impairments, children with NLD have good verbal skills, but a neuropsychological profile characterized by impairments in nonverbal abilities (Rourke, 1995). One of the identifying features of NLD is a significantly higher performance on tasks measuring verbal intelligence than on those measuring visuospatial intelligence (Cornoldi et al., 2003, Johnson, 1987, Mammarella et al., 2009 and Weintraub and Mesulam, 1983). A factor underlying this discrepancy is that children with NLD possess poor visuospatial and visuoconstructive abilities, which would explain their difficulties in a broad range of school and everyday life activities including, but not limited to, mathematics, drawing, and spatial orientation (Cornoldi et al., 1995, Cornoldi et al., 1999, Cornoldi and Vecchi, 2003 and Mammarella and Cornoldi, 2005). Most important for the current purposes, children with NLD often manifest emotional and relational difficulties that prevent them from adequately processing emotional information (Petti et al., 2003 and Rourke, 1995; see also Worling, Humphries, & Tannock, 1999). Finally, it has been found that children with NLD suffer from several impairments in social problem-solving, suggesting a reciprocal influence between socio-cognitive factors, the development and maintenance of pathological symptoms, and adaptation to the environment (Galway & Metsala, 2010). In summary, children with LD are an important population to be considered in the study of memory illusions. Several behavioral patterns could emerge from our research: one could expect a general poorer performance in recognition memory in these populations of children. Alternatively, memory performance, particularly memory illusions, may manifest differently within these unique groups of children with disabilities, depending on the purported mechanisms underlying the illusion. 2. The present study In the present study, children with NLD, children with dyslexia, and typically developing children were administered a false-memory task adapted from a paradigm previously used with typically developing children (Lyons, Ghetti, & Cornoldi, 2010) and children with ADHD (Mirandola et al., 2012). This particular paradigm allows emotional false memories to be investigated in both children and adults (Mirandola, Toffalini, Grassano, Cornoldi, & Melinder, 2013). Participants first study photographs that depict common actions for a variety of commonplace daily scenes. For each script, participants view typical scene-appropriate actions, as well as photographs that depict the effect of an action whose cause is not presented in the episode. Specifically, this paradigm allows for the investigation of two types of memory errors: gap-filling and causal errors incurred when one falsely recognizes, respectively, scripted information that was not presented (e.g., eating some food at dinner), or causal information (recognizing a cause photograph, i.e., “knocking over a bottle of water on the table” when they only viewed the effect of that cause, i.e., “pieces of a broken bottle on the floor”). The two main objectives of the present research were: (1) to investigate whether children with learning disabilities are less prone to memory errors compared to typically developing children (as it has been observed with the DRM paradigm), specifically when scripted episodes that depict real-life events are presented at encoding, and (2) to evaluate whether differences in false memory formation could emerge as a function of developmental disability. Specifically, we expected that children with NLD would attend to the negatively charged events to a higher degree than their peers, due to their vulnerability to emotional distress, and would thus be more prone to evince false recognition for negatively charged episodes. This prediction is based on the foundation that memory errors are most likely to occur when the error is consistent with the semantic gist or theme of the studied material (Brainerd & Reyna, 2005; Howe, 2007), as in the case of script consistent non-presented lures (i.e., gap-filling error). When individuals erroneously attribute causes to experienced effects (i.e., inferential causal error), emotional involvement may be critical; stimuli with emotional valence have been argued to elicit gist connections more than stimuli of neutral valence (Brainerd, Stein, Silveira, Rohenkohl, & Reyna, 2008). Increased false recognition of negative actions is not expected in children with dyslexia; their overall reduced semantic elaboration abilities was expected to result in overall lower false recognition for causal inferences and gap-filling errors regardless of the valence of the stimuli. Finally, in the present study we wanted to test the subjective quality of the children's memory experiences. We did so by asking children to generate Remember/Familiar judgments for each recognition decision, given initial evidence that children with developmental disabilities have different subjective memory experiences than their peers (Mirandola et al., 2011). Recollection and familiarity have been previously investigated in children with severe developmental disabilities (see Costanzo, Vicari, & Carlesimo, 2013 and Bigham, Boucher, Mayes, & Anns, 2010, for Williams’ syndrome and autism respectively), but it seems an important research avenue to explore in children with milder learning difficulties. It is possible that children with milder LD may display reduced subjective experience of recollection for both true and false recognition rates due to general impairment in semantic elaboration processes. This prediction is based on past findings in adolescents with reading comprehension difficulties that reported lower recollection rates than a control group, consistent with general memory impairments in children with learning disabilities (Mirandola et al., 2011). However, differences could emerge depending on the type of disability. Specifically, we predicted that children with NLD would display a higher tendency to experience subjective recollection for emotionally charged situations, given their often reported higher emotional involvement. Conversely, we expected a reduced subjective experience of recollection linked to both true and false memories in children with dyslexia, given their likely lower ability at handling conceptual information.