Using Recognition Testing to Support Semantic Learning in Developmental Amnesia

Abstract

Patients with developmental amnesia (DA) have suffered hippocampal damage in infancy and subsequently shown poor episodic memory, but good semantic memory. It is not clear how patients with DA learn semantic information in the presence of episodic amnesia. However, patients with DA show well-developed recognition memory and these recognition abilities may support semantic learning. We present data from three experiments (two previously described in Elward & Vargha-Khadem, 2018). The first experiment showed that recall tests did not facilitate semantic learning. Patients with DA recalled only 35% of the learned information (controls recalled 80%). The second experiment indicated that multiple-choice recognition tests may facilitate learning. Patients with DA recalled 85% of the learned information. In experiment three, a patient with DA (aged 8 years) took part in a repeated-measures test so that recall learning and recognition learning could be directly compared. The results showed a clear benefit of recognition learning compared to recall learning (76% v. 35%). This finding indicates that young people with extensive hippocampal damage indeed utilise their recognition memory to support the integration of new information into their semantic system. This has important implications for the support of school-aged children with episodic memory difficulties.

Episodic memory development: Bridging animal and human research

“… 10 Furthermore, their deficit in memory retention seems to be mediated via an inability to retrieve memories via (autonoetic) recall. If memory retention is assessed via recognition memory tests (e.g., multiple choice), DA patients often perform comparably to controls 112 , 113 (see Box 2 for a further discussion of DA and its implication for developmental neuro-plasticity). Thus, the hippocampus seems critical for the development of allocentric what-where and W-W-W encoding, as well as recall, while the development of what and egocentric what-where memories, particularly when accessed via recognition processes, may be less dependent on the hippocampus.…”

Resetting the Hippocampal Buffer: A Neurocognitive Account of Psychedelic Therapy for Anxiety-Related Psychopathology

“…If extinction learning is enhanced by the acute effects of psychedelics (Cameron et al, 2018;Kelly et al, 2023;Rogers et al, 2024;Werle et al, 2024;Woodburn et al, 2024) and such learning is hippocampal-independent, then exposure to anxiety-related cues during acute effects may produce a context-general extinction memory. Likewise, if cortical learning during acute effects is enhanced (Doss et al, 2023), presenting information that could disrupt anxietymaintaining beliefs might be warranted, especially with repetition given that the cortex learns with repetition (also note that extinction involves repeated presentation; Elward et al, 2023;Elward & Vargha-Khadem, 2018;Verfaellie et al, 2008;Yonelinas, 2002). Finally, as the acute effects wane and hippocampal function ostensibly normalizes, it may be crucial to focus hippocampal encoding on information inconsistent with an anxiety state to promote a sustained suspension of the maladaptive cycle of anxiety-related hippocampal-cortical constraints predicted by this model.…”

Developmental Sequences Constrain Models of the Mind

“…Research with developmental amnesia patients shows that repeatedly encountering material leads to them to develop a feeling of familiarity, as tapped by recognition testing. Familiarity might support the acquisition of semantic information (Elward et al, 2023) although as with indirect assessment of statistical learning, this idea requires evaluation. It is also possible, as Forest and her co-authors suggest, that infants may form very short-lived episodic memories that might support the development of gist memory, but the evidence for such memories is scanty .…”