This application of statistics before heuristics marks a departure from traditional hand coding and text-processing techniques.
Underpinned by empirical, self-emergent trends as identified across the broad Autism corpus, these emerging patterns enable discussion as to the past, current and future direction of Autism research, within a broader context of new health initiatives.
Specifically, the a priori application of user-imposed heuristics to delineate expected thematic clusters, coupled with the subjectivity of hand assessment, artificially constrains and shapes results, rather than allowing self-emerging patterns to be identified.
This article seeks to empirically explore and examine whether broad trends toward a biological and physiological research focus of ASD have continued in light of the consolidation of disorders under the ASD umbrella and recent health initiatives, when no a priori expectations or limitations on thematic trends are imposed.
Combining traditional bibliometric co-word techniques, with tenets of graph theory and network analysis, this article provides an objective thematic review of research between 19 to consider evolution and focus.
Results illustrate growth in Autism research since 2006, with nascent focus on physiology.A total of 25,782 extracted keywords were pre-processed to trim white space and remove numeric only entries before being converted to lower case and stemmed to characters 1:6 (facilitating the automatic augmentation of similar terms, e.g., Behavior, Behavioral, and Behavioral Intervention were subsumed into a single representative term).Duplicates were combined creating a unique dictionary of 6242 stemmed keywords for corpus analysis.An initial frequency-based time-series assessment was applied to identify the publication rate per year, the most frequently occurring journals for publication per year, and summative citation analytics.Unfortunately, the provision of author keywords is notably inconsistent—with many failing to provide such text markers—as such, database attributed keywords were extracted to isolate areas of research focus.Bibliometric information was extracted for each article, including author(s), affiliation(s), publication title, journal, attributed keywords and cited references.This information was formatted and processed within R software (R Core Team, 2016) using functionality within the ‘tm’ and ‘Bibliometric and Co-citation Analysis’ packages (Aria and Cuccurullo, 2017).This reconceptualization was completed with the publication of the DSM-5, which marked the consolidation of previous diagnostic terminology (American Psychiatric Association, 2013; see Figure 1A).However, the amalgamation of formerly heterogeneous disorders under a single diagnostic term was, and arguably continues to be, contentious (Singh, 2011).Autism spectrum disorders (ASD) is an umbrella term encompassing the diagnoses of autism, Asperger’s syndrome (AS) and pervasive developmental disorder – not otherwise specified (PDD-NOS) (American Psychiatric Association, 1994; American Psychiatric Association, 2013).Proposed in the early 1980s (Wing, 1981, 1997; Nordin and Gillberg, 1996), the ‘Autism Spectrum’ was first clinically conceptualized with the publication of the fourth edition of the Diagnostic and Statistical Manual of Mental Health Disorders (DSM-IV, American Psychiatric Association, 1994).