|
Perspegrity Solutions -- Increasing the quality of connection and decision making in online work, learning, and community -- |
Example features supporting ethical thinking skills
- Supporting multiple perspectives
- Supporting task accountability
- Articulating certainty levels
- Transparency in on-line moderation
- Explicit modes of inquiry
Below are descriptions of several simple software features within the "technology supporting ethical thinking skills" theme. A longer list of possible features can be found here. While these examples highlight single illustrative features, real systems will usually incorporate a number of such features. (These features do not make use of sophisticated or novel technologies, as these are not needed to illustrate the approach, but advanced methods from research in adaptive systems, machine intelligence, collaborative filtering, etc. could be incorporated to make them more effective.)
As I describe these software features I will mention some "general principles" that serve as base assumptions and as reasonable hypotheses to be tested empirically.
1) Supporting multiple perspectives. Imagine an adult learning class in which the instructor places a high value on students considering multiple perspectives and alternative viewpoints. The instructor uses a template-based tool (like WebCT) to create web pages for the class. This web site design tool has a simple feature to support student's consideration of multiple perspectives. Every web page with content has a button/link in upper right corner titled "Other perspectives and alternative ideas" which leads to additional content. Depending on the context, the link might lead to competing scholarly theories, alternate ideas authored by other classmates, etc.
General principle #1: Salience without use. The visual presence of the feature itself can have an effect, even if it is not used.
Students are not forced to use this link. However, its constant visual presence serves as a reminder that, whatever they are reading, there might be (and probably are) other perspectives on the issue.
General principle #2: Salience-- usage--internalization. Feature salience supports feature use; which in turn supports an internalization of the intended behavioral/thought habit.
The addition of a simple technology feature supports learning by providing clear and persistent cues or structures. For example, the existence of the multiple perspectives feature will lead to some degree of use, which will lead to some degree of increase in students thinking/communicating with an awareness of multiple perspectives. The effect may of course be small, in part because general thinking and communicating skills develop slowly (and see principle #3).
General principle #3: Need for social/contextual supports. Individuals in the "learning community" will need instructional support and models of proper use. They will need feedback after use and will need to discuss new tools and norms among themselves to fully understand the tool's purpose. Practices and procedures must be adopted to accompany the new tools.
A technological feature by itself is not likely to produce change or learning without sufficient support in the social system (Wenger 1998). For example, with the "multiple perspectives" feature above, the instructor might discuss multiple perspectives in class and have a class discussion about how students used the multiple-perspectives link feature.
2) Supporting task accountability. Suppose that, due to pervasive inefficiencies in getting things accomplished, an organization decides to start placing a higher value on responsibility and accountability in members' communications. Leadership sets a goal that participants will clearly specify the person responsible and the date of completion for any action plan. Members may have only sporadic success in reminding each other to adhere to these new guidelines in private conversations, in meetings, and in memos. But it is possible to modify the group's email software such that email messages contain a template for specifying the responsible person[s] and the date due (similar to the action workflow approach described in Winorad & Flores 1986).
As indicated in Principle #1, this feature could be ignored, but its existence would provide both a constant reminder to include this information, and a structured consistent place to enter and find this important information. (We could even imagine, though the image may not be very attractive, a computer program that inspects all internal email traffic to create lists of responsibilities and reminds people to enter this important information if they fail to do so in their emails.) And from Principle #2 we can imagine that, through the repeated reminding and use provided by electronic interactions, members of this organization might internalize the desired behaviors and values and learn to specify responsible persons and due dates in their off-line interactions.
We need not debate the merits or the oversimplification of this example--it is given only as another illustration of how a simple technological feature can serve to cue and structure ethically-oriented behavioral/thinking habits.
3) Articulating certainty levels. Imagine a community that regularly uses an on-line forum to discuss important issues. The discussion forum is a typical one except for the following additional feature: when one posts an idea or a response, one can select from a menu of four choices to describe their "certainty" in their idea. The choices are, from most to least certain: "fact/observation/data," "opinion/inference," "guess/hypothesis," and "question/request." Users are not required to use this feature, but the group's stated policy encourages its use. This feature encourages a type of metacognition and epistemic sophistication (Murray & Ross 2006). A user who types a strongly opinionated post is thus encouraged to pause and reflect on whether his idea is actually his opinion as opposed to a hard fact. Tagging a strong statement as an "opinion" can tempter its emotional intensity. The author benefits from the reflective moment, and the reader benefits because, for example, one does not react as strongly to a communication that was explicitly (just) an opinion. If used extensively in a community of learning, the associated reflective skills may improve.
Many online forum tools include the ability to label forum postings in this way, and the labeling categories mentioned above were used in a community forum that I was involved in creating. Anecdotal reports from users indicated that the feature had its intended effect of easing tensions for both author and reader for controversial discussions.
Would the extra effort of marking all of one's postings with a confidence rating be worth it in terms of added communicative benefits? What type of participants have the capacity to rate their confidence in a way that is consistent and accurate? These questions can only be answered through scientific trials.
General principle #4: Developmental limits. There are developmental constraints on the complexity and abstraction of people's reflections (Kegan 1994). These may limit the usefulness of features supporting ethical thinking skills in some contexts.
4) Transparency in on-line moderation. Imagine again a community that regularly uses an on-line discussion forum, but this time the community is more amorphous and in constant flux. Such forums are often rife with "junk" postings, and need a moderator to act as a gate-keeper. Forum moderators have tremendous power to influence the nature of the on-line discussion in their selection of what messages to delete. Here is an example of a feature that, through moderator transparency and accountability, mitigates such abuse of power and enhances group trust: the moderator does not "delete" postings, but instead they move them to one or more peripheral folders that are open for inspection. (Example names for such folders might be "profanity," "spam," "flaming," and "too tangential".) The moderator's choices are then public.
General principle #5: Increased cognitive and task load. The addition of features that encourage participants to "go meta", i.e. reflect on, be explicit about, or discuss truth, certainty, biases, group process, etc. could add both cognitive load from the additional depth of thinking, and add content load with the extra discussion engendered.
Excessive thinking about thinking, dialoging about dialog, or theorizing about theories can sabotage any collective enterprise (so-called "analysis paralysis"). For example, the moderator transparency feature above creates an opportunity for protracted discussion about the validity of the moderator's choices. Also, the "confidence levels" feature described in example #3 above could instigate lively philosophical discussion about when an idea can be said to be an objective "fact" or "observation." The proliferation of such extra discussions may or may not be aligned with the group's goals, and an appropriate balance between content and meta-content (or meta-dialog) must be found in each situation. As with most innovations, adopting the tools described here is not without risks and tradeoffs.
5) Explicit modes of inquiry. Imagine a knowledge base or web site that includes information on a variety of topics, some of which are controversial. The software used to create this web site allows any paragraph (or any item of information) to have a button (link) to "Validation Information" placed next to it. Users would click on this link to read information that would help them determine the degree of trust, certainty, or validity of that item of information.
General principle #6: Epistemological concerns such as the validity of information in a knowledge base and the modes of validation that people use to agree upon shared knowledge have strong interactions with ethical concerns.
Note that the system does not have to bias any particular validation mode. A scientist can find (or add) supporting or refuting information based on empirical work, and a religious fundamentalist can find (or add) supporting or refuting information based on the authority of scripture. But both of them are supported in reflecting on how they validate information and the fact that they use entirely different validation modes, which could help them clarify their differences and engage in more productive dialog
THIS SITE IN BETA -- Content
on this site is licensed under a Creative
Commons Attribution 2.5 License