The free rider problem and tradegy of the commons (Fisher, 2005)

24 October 2010

There are two classes of social dilemmas especially relevant to information behavior: the free rider problem and the tragedy of the commons. These dilemma situations refer to opposite sides of the same coin. In the free rider problem actors are tempted to not contribute to a group good, while in the tragedy of the commons actors are tempted to consume a good without consideration of how their use degrades that good. The free rider problem arises when actors want to enjoy a collective good, without contributing the resources necessary to create or maintain it. The seminal statements come from extensions of the rational actor model (derived from economics) to political and social problems of collective action (Olson, 1965), theories of group solidarity (Hechter, 1987), and the enforcement of norms (Coleman, 1990). The free rider problem is seen as a pervasive challenge for any collective good providing group, and is especially problematic when contribution is voluntary and groups are informal. This is often the case in online groups, so studying provision of collective goods in these settings promises to bring new insight into this general question.

The tragedy of the commons (Hardin, 1968; Ostrom, 1990) occurs when individual consumption of a collective good degrades the quality of that resource for all. When the resource is large none of the individuals feel the negative effects of their own actions, thus they have no incentive to curtail their own consumption (Yamagishi, 1995). The tragedy of the commons highlights different ways that consumption of a resource interferes with the ability of others to use that resource. Bandwidth consumption due to excessive downloading is a clear example in the online setting (Huberman, Rajan, &. Lucose, 1997). However, one of the major strengths of electronic resources is that they greatly reduce problems of rivalness (only one person can read a book at a time), degradation (Web pages don’t wear out), and crowding (for most uses additional users do not noticeably degrade the quality of online information resources). However, there are online social spaces like Usenet groups, e-mail lists, and blogs that may have a social carrying capacity where excessive use by some may degrade the resource for others.

The baseline prediction in collective action dilemmas is that the deficient equilibrium will dominate in the absence of mechanisms that alter the payoff structure. Identifying and explaining the operation of such mechanisms has been the subject of a wide range of research. Contribution, cooperation, and trust are more likely to emerge when actors know that they will interact in the future, i.e., repeat games cast a “shadow of the future” (Murnighan &. Roth, 1983). Reputation effects are reason this shadow effects dilemma situations. Actors “do the right thing” in order to protect their reputation, an important mechanism for generating trust in online settings (Kollock, 1999). In-group membership due to identities based on roles or on group affiliation can foster contribution when embedded in social ties (Snow, Zurcher & Exland-Olson, 1980). Social ties, that is, the connections between individuals, are an important reason that people contribute to collective goods (McAdam & Paulson, 1993). Finally, selective incentives, in the form of valued goods (especially social incentives like approval), are only available to those that contribute and are important foundations for sustaining contribution (Coleman, 1990; Hechter, 1987). In the absence of the mechanisms like those described above, cooperation and contribution are harder to sustain, and are more likely to devolve into defection.

(Fisher, 2005, p. 94ff)


Affective Load Theory (ALT), (Fisher, 2005)

24 October 2010

Affective load theory (ALT) is a social-behavioral perspective on the thoughts and feelings of individuals while engaged in information behavior (lB). ALT provides empirical methods for identifying affective states of users that disrupt ongoing cognitive operations (James & Nahl, 1986). Once a disruptive affective state is identified, coping assistance services (CAS) can be provided to encourage users to mitigate disruptive states to achieve task success. ALT identifies underlying habits of thinking and feeling while engaging in information behavior, and clarifies the details of information retrieval from a user perspective. There are three essential ideas in applying social-behavioral psychology to IB:

  1. The mental activity of information users, both cognitive and affective, is defined as behavior (Martin &. Briggs, 1986). For instance, “thinking of a search word” or “feeling motivated to finish a task” are behaviors. Global control of the affective over the cognitive operates at general and specific levels. At the general level of control people possess motivational states such as optimism or pessimism prior to a search. At the specific level of control we experience micro-behaviors that involve search strategy such as inspecting a list, thinking of a synonym, or recalling an item that has been seen before. A search task or session involves hundreds of individual cognitive micro-behaviors, each one connected to an affective state that maintains or interrupts it (Nahl, 1997). Affective states are organized in a top-down hierarchy and can be reliably measured through concurrent self-reports about expectations, satisfactions, and acceptance during continuous cognitive activity.
  2. Affective behavior initiates, maintains, and terminates cognitive behavior (Isen, Daubman & Gorgolione, 1987; Carver & Scheier, 2001). For instance, when searchers lose the motivation to continue a task, they begin thinking about something else. Or, if they unexpectedly find some new information they want, they switch activity midstream. The new affective behavior interrupts and takes over the ongoing activity and continues in a new direction with new cognitive activity. This managerial or directive function of affective behavior over cognitive, makes it desirable in information environments to employ self-monitoring techniques to keep track of the affective behavior of users (Nahl, 1996, 1998).
  3. Affective behavior operates within a binary-value system: on/off or positive/negative. Cognitive behavior operates through a multivalue logic. Therefore, affective behavior is measured with bi-polar scales and cognitive behavior is measured with multiple-choice, matching or fill-in items. Content analysis and protocol analysis of concurrent verbal reports are used to identify affective and cognitive behavior patterns during search tasks (Nahl, 2001).

The behavioral approach to information use is attracting increasing interest among information scientists. However, sufficient attention is not given to the three essential elements outlined above. The focus has been on cognitive behavior and more recently, on how affective behavior is also important to consider. Nahl’s social-behavioral theory of affective load makes explicit the need to create a methodological connection between each cognitive behavior and its affective support or control state. Affective load theory was developed by analyzing concurrent self-reports of searchers and learners in conjunction with quantitative ratings filled out by searchers while engaged in searching and problem-solving. To achieve high reliability, it is critical to obtain concurrent rather than recollected data. Nahl’s ALT theory is emerging from a 20-year research program. One area of application has been to identify affective dimensions like self-efficacy and optimism that help searchers perform better. Currently ALT research focuses on how diverse affective behaviors interact to produce an effective coping style when searchers feel challenged by uncertainty.

ALT proposes that all information behavior involves affective states that provide specific goal-directionality and motivation to support cognitive activity. Affective load ( AL) is operationally defined as uncertainty (U) multiplied by felt time pressure (TP). Uncertainty is defined as the combined degrees of irritation, frustration, anxiety, and rage (Nahl, 2004).

AL = U (irritation + frustration+ anxiety + rage) x TP

Affective load is high when people operate with ineffective cognitive behaviors. For example, cognitive ambiguity, uncertainty, or information overload attract affective behaviors that are negative and counterproductive to the searcher’s goal. For instance, a search that appears to yield no relevant results after some attempts is cognitively disorienting, as represented by such thoughts as, ‘Tm no good at this” or “This is so frustrating! ”

At other times, searchers are able to engage affective coping strategies when faced with cognitive load and uncertainty. For instance, “I’ll just keep going until I find something” or “I’m positive I can find what I need in another database.” These verbal expressions are standard and recurrent within a population of searchers, and because they are learned cultural habits, can be termed “learned affective norms” (LANs). Negative LANs disrupt cognitive strategies, interrupt the search, and often terminate it prematurely, while positive LANs provide persistence and integration to cognitive strategies. In general, negative LANs increase AL and appear in the form of uncertainty, anxiety, frustration, low expectations, pessimism, low self-efficacy, low task completion motivation, low satisfaction, low system acceptance, and other disruptive symptoms that interfere with a positive outcome. On the other hand, positive LANs decrease AL because they provide better coping strategies to manage ambiguity and cognitive load. Support and counseling interventions can be triggered when affective load rises above a specified level. Knowledge about the affective environment of searchers will also be helpful in search instruction. More research is needed on how the affective information environment of searchers impinges on their cognitive activity to strengthen information system services and design.

(Fisher, 2005, p. 39ff)