The Ladder of Inference: Making Informed Choices

The ladder starts with data. There is a tremendous amount of data in the world and we have a limited capacity to absorb it. Research suggests that we take in only about 10% of the data available. So, we select data. We then add meaning to the data we select in order for it to make some sense to us. Based on the meaning we add, we form conclusions and then take action based on those conclusions.

The cycle begins again when the results of our actions become data for another trip up the ladder.

Over time, our conclusions form our beliefs, assumptions, and values; these filter the data we select and the meaning we add when we repeat this process. This is another way of illustrating the point made in the last section: our mental models tend to be part of reinforcing, self-perpetuating structures.

Implications

Our capacity for making inferences is an essential skill, allowing us to act quickly in emergencies and efficiently in routine situations.

We assume that other people see the world as we do. When we disagree with others, we usually argue about our conclusions. We assume that we have selected the same data and added the same meaning to this data.

We usually "leap" up the ladder, unconscious of the rungs we've climbed.

We often make these leaps without testing our reasoning process or the data that started the cycle.

We assume that our conclusions are the "truth."

Our assumptions and beliefs are often privately held and not tested.

How to Use the Ladder of Inference

Conscious use of the ladder of inference is a simple way to increase collective intelligence. Experiment with the following steps in low-risk situations, then apply them in higher stakes situations. If you are accustomed to win/lose debates with some people, they may be suspicious of your new approach at first. Be patient.

Practice naming assumptions you are making and their influence on your actions.

Play "detective" or "defense attorney" and insist on separating the data or evidence from the conclusions. Notice the types of conclusions you tend to mistake for the facts.

When making recommendations (or advocating), share your "ladder" or reasoning.

When others make a recommendation, ask them to share their reasoning with you.

When you surface disagreements based on different data, ask those involved to assume that all the known data is "true." What new conclusions can be drawn? Is there other data that would have an impact on these conclusions? How can we expand our data pool?