Many theories addressing the complex anxiety-cognition interaction are built upon the notion that working memory is vulnerable to the effects of anxiety. However, most research has utilised studies of trait anxiety which does not allow direct inferences to be made between affect and cognitive performance, or exclude confounds such as pre-existing individual differences. As a result, a systematic review was undertaken to explore the neuropsychological consequences of experimentally-induced state anxiety on working memory. Twenty eight studies were included in the review grouped according to the nature of anxiety induced (anxious-apprehensionor anxious-arousal) and by working memory component (verbal, visuospatial, or executive). This review found evidence of anxiety adversely affecting working memory and support for specific theories. The review highlighted the potential impact of anxiety on neuropsychological assessments in clinical settings, as well as the need to test these theories using prolonged anxiety-induction procedures with standardised measures of anxiety and working memory. The empirical paper explored the effects of experimentally-induced anxiety on clinical neuropsychological assessments of working memory. Anxiety was induced using the inhalation of carbon dioxide (CO2), which reliably elicits prolonged states of anxiety. Thirty participants were included in the study, undertaking neuropsychological tests during the inhalation of both normal air and CO2 across two testing sessions using a counterbalanced design. The mood manipulation was successful and verified using psychological and physiological measures. Results suggested that anxiety negatively affected spatial and verbal working memory, although low load verbal tasks were unaffected. Clinical implications of these findings are discussed with regards to the effect of state anxiety on clinical neuropsychological assessments and the management of anxiety in light of these cognitive deficits. Future research should include a broader spectrum of working memory tasks of varying cognitive load to further elucidate how anxiety may interact with cognition.