This thesis investigated the memory skills of young adults with and without language and learning disabilities (LLD) using the Deese-Roediger-McDermott word recall paradigm (Roediger & McDermott, 1995). Three types of word lists were presented: semantic lists consisted of words that are related to a non-presented critical item (CI) (e.g., bad) in meaning (good, rotten, harmful, worse); phonological lists included words related to the CI in sound (e.g., had, lad, bat, bag); and hybrid lists included words related to the CI in both meaning and sound (e.g., good, lad, rotten, bat). Individuals with diagnoses of LLD were classified as "true LLD" or "compensated LLD" based on language test scores and a discriminating composite score, while those without LLD were considered the "typical language" (TL) group. Hypotheses were made regarding veridical recall and memory intrusions, including intrusions of the non-presented critical item (CI). For veridical recall, the compensated LLD and TL groups were expected to have higher recall accuracy than the true LLD group. As for CI intrusions, two possible outcomes were considered: the true LLD group may recall more CIs due to inability to discriminate between presented and non-presented words (Kirchner & Klatzky, 1985); or they may recall fewer CIs due to difficulties forming traces of the gist of the word list (Weekes et al., 2005). Data from 30 participants (ages 18 to 25) -- 12 true LLD, 8 compensated LLD, and 10 TL -- were included in this thesis. Results indicated that the true LLD group showed a non-significant tendency to have lower recall accuracy scores than the other two groups, and a higher number of CI intrusions. List-type also affected accuracy and CI intrusions, as semantically-related lists increased recall accuracy and hybrid semantic-phonological lists increased CI intrusions. Possible conclusions from these data are presented along with recommendations for future research. / text