Forgetting in short-term memory has generally been attributed to the passage of time or interfering events. Various models tracing patterns of recall over brief periods of time ascribe forgetting to decay of information with time, loss of distinctiveness of information, or interference from other material. This study attempted to choose between the Primacy model, the SIMPLE (scale-independent memory, perception, and learning) model, and the SOB (serial order in-a-box) model that would best explain performance on a visual serial recall task manipulating delay at encoding and retrieval. The results failed to find detrimental effects of decay with longer intervals, refuting predictions of the decay-based Primacy model. In contrast, an interference-based SOB was found to reasonably explain the results obtained. Analysis of error patterns revealed the Primacy model to not best explain the error patterns noted in the data, contrary to popular findings in serial recall literature. The SIMPLE and SOB provided satisfactory explanation for error patterns, with the SOB accommodating error patterns better. It can be concluded that purely temporal models of forgetting cannot offer adequate explanation for forgetting in visual STM.