Table 2.
Active technology role and theoretical grounding.
| Reference and project or intervention name |
Active technology type | Dynamic tailoring | Interactive education | Self-monitoring | Theoretical grounding |
| Ananthanarayan & Siek 2010 [27] (HealthSense) | Inference; pattern recognition | Not specified | Yes, but details not given | Yes; provide awareness of physical activity | General awareness only; no specific theory mentioned |
| Arteaga et al 2009 [28] | Dialog; pattern recognition | Not specified. Static tailoring only | No | Not specified | Big 5 personality theory; technology acceptance model; theory of planned behavior, theory of meaning behavior |
| Bickmore & Picard 2005 [29] (FitTrack) | Dialog | Not specified | No; passive educational content only | Very basic, pedometer steps only | Relational agents |
| Bickmore et al 2005 [30] (FitTrack) | Dialog | Not specified | No; passive educational content only | Progress charts only | Relational agents |
| Bickmore & Sidner 2006 [31] | Inference; dialog | Not specified, but possible | No | Progress charts only | TTMa, MIb: link with agent reasoning and ontology |
| Bickmore et al 2009 [32] | Pattern recognition | Not specified in detail | No | No | Relational agents |
| Bickmore et al 2009 [33] | Dialog | Not specified in detail, only mentioned as a property of dialog systems in general | Yes, support low health-literacy patients | No | Relational agents |
| Bickmore et al 2010 [34] | Dialog | Not specified in detail | No | Not considered usable by schizophrenia patients | Relational agents |
| Bickmore et al 2011 [35] | Inference; dialog; user models | Not specified; fixed tailoring only mentioned | No, but mentioned in a generic way | No | TTM, MI encoded in ontology for agent reasoning and user model |
| Bickmore et al 2010 [36] | Dialog | Not specified | No | Charts only | Relational agents |
| Bieber et al 2009 [37] (DiaTrace) | Physical activity recognition, mobile phones | Not specified | No | No | Not mentioned |
| Buttussi & Chittaro 2008 [38] (MOPET) | Pattern recognition; adaptation; user model | Yes, due to context awareness | No | Not mentioned, but possible | Not mentioned |
| Consolvo et al 2008 [39] (UbiFit) | Activity recognition; inference | Not specified | No | Yes; visual display | No |
| De Rosis et al 2006 [40] | Dialog; user modeling, adaptation | Yes, due to adaptation | No | No | TTM |
| Farzanfar et al 2007 [41] | Dialog; pillbox sensors + adherence data analysis—linked to dialog system | Not specified, although possible | Yes, telephone instructions but limited interactivity | No | Self-efficacy theory, MI |
| Hakulinen et al 2008 [42] (COMPANIONS project) | Dialog; automated planning; knowledge-based inference | Not specified, although possible | No | No | Not mentioned |
| Hartmann et al 2007 [43] | Inference: evidence-based decision rules | Not specified in detail, but possible | Yes, but limited | No | No |
| Hayes et al 2009 [44] | Context-aware reminders; activity recognition; rule-based inference | Yes, decision to prompt based on recognized activity pattern | No | Not mentioned, but possible to include | Not mentioned |
| Jin 2010 [45] | Virtual agent in game | Not specified | Yes, education-entertainment | No | Health belief model, self-efficacy |
| Kaipainen et al 2011 [46] | Context awareness, pattern recognition, inference, planning, user modeling | Yes, messages tailored to changing context of user | Not a main focus | Not mentioned, but possible to include | Hybrid approach including self-efficacy and social influence |
| Klein et al 2011 [47] (eMate) | Knowledge-based reasoning; user models | Yes, automated reasoning based on COMBIc model ensures dynamic tailored messages depending on user’s context and state of mind | No | No | COMBI model includes aspects of TTM, health belief model, social cognitive theory, self-regulation theories, attitude formation theory, and relapse prevention model; interaction based on MI |
| Konovalov et al 2010 [48] | Pattern recognition; inference | No, but could be used in an intervention with dynamic tailoring | No | No | No |
| Lee et al 2010 [49] | Pattern recognition; user modeling (profiling), including mental states | Not specified in detail, but planned | Not specified, but planned | Not specified, but planned | Action-based behavior model: (1) increase user’s awareness of health; (2) set goals; (3) educate user in how to achieve goal; (4) remind; (5) reward + assess |
| Levin & Levin 2006 [50] | Voice recognition; semantic representation; dialog | Not specified, but personalization of dialog possible | No | No | No |
| Lisetti & Wagner 2008 [51] (ABLE) | Dialog system considered | Not specified, but possible | No | No | MI |
| Looije et al 2010 [52] (SuperAssist) | Dialog | Not specified, but possible | No | No | MI |
| Maier et al 2010 [53] (SEMPER) | Text mining; ontologies; machine learning; semantic search | Yes, personalized search results based on user profile built automatically | Yes, information portal | No | MI |
| Mazzotta et al 2007 [54] (PORTIA) | Dialog, user model | Yes, tailoring of persuasion messages based on inferred personality traits and likely motivations of user | No | No | Persuasion theories, argumentation |
| Munguia Tapia 2008 [55] | Activity recognition; energy estimate | No, but possible in an intervention | No | No, but possible in an intervention | No |
| Nguyen & Masthoff 2008 [56] | Dialog | Not specified | No | No | MI-based dialog design |
| Oddsson et al 2009 [57] (SKOTEE) | Intelligent reminding | Yes, part of robotic companion | No | Not mentioned, but possible to include | No |
| Op den Akker et al 2011 [58] | Pattern recognition, machine learning, context awareness, user modeling | Yes, messages are tailored based on user model and context | No | Not mentioned | No |
| Rojas-Barahona & Giorgino 2009 [59] (AdaRTE) | Dialog; adaptation | Yes, dialog can be adapted according to patient answers | No | No | No |
| Smith et al 2008 [60] (COMPANIONS) | Dialog control; inference; automated planning | Yes, update planned activities through ongoing dialog | No | No | No |
| Sorbi et al 2007 [61] | Adaptation, automated personalized feedback | Yes, tailored messages depending on current experience | No | No | No |
| Spring et al 2010 (Make Better Choices–MBC) [62] | Decision support; coaching algorithms. (PDAd) | Not specified, but possible | No | Yes, PDA allows this but not described in detail | No, although some theories mentioned |
| Tiwari et al 2011 [63] | Robot, dialog | Not specified in detail, but dynamic adaptation is a required feature in the design | No | No | No |
| Turunen et al 2011 [64] (COMPANIONS project) | Dialog; inference; automated panning | Yes, adaptive dialog, collaborative planning | No | No, but possible to include | No |
| Uribe et al 2011 [65] | Knowledge-based inference | Yes, reminders based on inferred mental state | No | Yes, implied in the design but not described in detail | TTM incorporated in ontology |
| van der Putten et al 2011 [66] (SERA project–Social Engagement with Robots and Agents) | Robot, dialog | Not mentioned | No | No | Not mentioned |
| Watson et al 2012 [67] | Dialog | Yes, dialog utterances tailored according to user progress with system | Not specified in detail | Not specified in detail | Relational agents |
a Transtheoretical model.
b Motivational interviewing.
c Computerized behavior intervention.
d Personal digital assistant.