Practical analysis for investment professionals
30 September 2019

How to Read Financial News Redux: Prevent Memory Contamination

How often do you say, “I read it somewhere”?

This is what happens when we remember facts but can’t remember where we learned them. This is a form of source amnesia, according to neurologist Steven Novella. Source amnesia is normal: We often acquire information but forget when and where we learned it.

Defining Memory Contamination

Our memories form over time as we recall events and create narratives that make sense to us. Our memories are usually reliable, but Novella points out the limitations:

“Our memories are not an accurate recording of the past. They are constructed from imperfect perception filtered through our beliefs and biases, and then over time they morph and merge. Our memories serve more to support our beliefs rather than inform them.”

Memory contamination happens at the time our memories are formed, both initially and subsequently. And if we read both fact and fiction, our memory will have a hard time separating the true claims from the false ones. This is called truth amnesia.

For example, let’s say we hear a rumor that apples cause cancer. We store it in our memory along with everything else we know about apples. And later, when our brains retrieve information about apples, the truth is contaminated by rumors and half truths.

This combination of source amnesia and truth amnesia creates memory contamination. Novella describes this process in Your Deceptive Mind: A Scientific Guide to Critical Thinking Skills. Memory contamination is not a problem with our ability to recall information. Rather, the flaw lies in how our original memories are formed.

Subscribe Button

Imagine you spend a day playing softball with friends and follow it up with dinner and conversation. Everyone has an opinion as to why the winning team won the game. Maybe it was the pitching, hitting, or defense. No one can remember every single play, but nevertheless, you create a narrative in your mind.

Meanwhile, as a social creature, you also want to conform your memories to those of other players. So as you discuss the game over your meal, you form a narrative — a memory — that is based on and influenced by those around you as well as their recall of the game, not on the game itself. And that narrative is not actually true.

This does not have to be deliberate deception, but we wind up with a narrative that is misleading, incomplete, or oversimplified.

So what does this have to do with investing?

In my last piece in this series, I described how Metallgesellschaft influenced oil prices in 1993 and 1994. No doubt journalists heard rumors about MG since it was a huge scandal at the time. But these rumors would have been a mix of facts, speculation, and innuendo. The rumors offered no clear and obvious narrative about oil prices. And they clashed with the longstanding understanding that behind every oil price movement lurked the omnipresent hand of the Organization of Petroleum Exporting Countries (OPEC). To counter the OPEC narrative would have meant going out on a limb and contradicting conventional wisdom — and divulging a truth that was not so easily understood.

In these situations, we wind up with a story that is simplistic and with a memory that is misleading.

This is more common than most investors would like to admit. Our memories of financial news and history are probably contaminated with all kinds of inaccurate or incomplete narratives.

And if our memories are contaminated, so are the rules of thumb we use to make investment decisions.

You could say that we suffer from “heuristic contamination.”

Tabloid Journalism

As if this isn’t challenging enough, the media caters to our need for simplicity. We want everything reduced to an easy-to-understand headline, preferably one that confirms what we already know. And the more sensational, the better.

These headlines mix fact with fiction to grab our attention and tend to reduce complex issues to simple cartoons. They are composed in two easy steps:

  1. Simplify
  2. Exaggerate

After you create a cartoon version of events, just inject a dose of fear into the mix and voilà! You now have a blockbuster article, ready to go viral.

Let’s say that someone claims that “Corporate earnings are crashing!” This triggers an emotional response that reinforces the memory, regardless of its accuracy.

Repetition also reinforces memories: If lots of sources say that “Earnings are crashing,” that will stick with us and we’ll be more likely to believe it is true. This is the foundation of political propaganda: Repeat the dogma until it becomes the truth.

We must be careful with sensational news because this nonsense can plant phony ideas in our heads, like weeds, that eventually creep into our long-term memory.

How to Reduce Memory Contamination

There is no easy way to develop a deep understanding of a topic. In-depth research requires that we consider a variety of perspectives. One way to reduce the risk of memory contamination is to inform ourselves with facts before exposing ourselves to the tabloid version. When we have a fully informed opinion and a broad understanding, sensational headlines have less of an emotional impact.

Here’s how to protect yourself:

  1. Start with the facts.
  2. Form your own opinion.
  3. Check other sources that you know are reliable (in case you missed something).
  4. Then you can be safely exposed to tabloid news without fear of contamination.

By the way, we can safely skip step four. Personally, I don’t think it’s healthy to read too many sensational headlines. I find that these exaggerations can get mixed into my memory and mess things up.

Even when we are aware of memory contamination, we are still vulnerable to its effects. It’s like dieting: Knowing about temptation does not make you immune from the lure of a hot fudge sundae. So you’re better off if you just steer clear of Häagen-Dazs and Baskin-Robbins.

Likewise, investment knowledge is not enough: You must change your investment process and your reading habits to mitigate your behavioral biases.

Putting It All Together

When I reviewed corporate earnings in early 2019, here are the steps I followed to understand consensus, question the narrative, and avoid memory contamination:

Corporate Earnings in Early 2019

1. Understand Consensus

FactSet Insight, 8 February 2019. This is my source of data about trends in corporate earnings.

2. Form My Own Opinion

Revenue growth in the fourth quarter of 2018 was inflated at 13.3%.

  • The energy cycle is boosting aggregate earnings: Revenue growth of 98% in 4Q inflates the aggregate revenue growth for the S&P 500.
  • The communications services sector has exaggerated revenue growth: Alphabet/Google was double-counted, and the 2018 numbers include traffic acquisition costs, boosting the sector’s revenue growth from 12% to 20%.
  • Looking ahead, revenue growth in 2019 should normalize at 5%, with 5% earnings growth.

3. Question the Narrative

  • Fundamentalis, various posts in early 2019.
    • Brian Gilmartin, CFA, confirmed the impact of the energy sector on earnings trends.
    • He also noted that Apple is dragging down earnings estimates (from 4.9% to 4.2%).
  • Dash of Insight, 9 February 2019
    • Jeff Miller didn’t raise any red flags about earnings in his weekly review.
    • People believe a false narrative that crime is rising in the United States.

Conclusion: My final opinion is unchanged, but I gained some additional depth and insight — without memory contamination.

Why Is It So Hard to Be Simple?

Back in 2015, Jason Voss, CFA, asked, “Which Challenges Do You Face Most Often in Your Professional Life?” The overwhelming majority of poll respondents cited information overload.

There is too much investment data out there, and we feel compelled to consume it all.

Why do we feel the pressure to read so much?

Sometimes we read to show how much we know, to impress clients and co-workers. Sometimes we read because office politics dictate that we keep up appearances. And sometimes we read because we are trapped by our own pride and hubris, believing that we will make better decisions if we read until our eyes turn red.


Reading financial news does not have to be so stressful. If I had to boil it all down to a few steps, I’d say:

  1. Understand what clients need.
  2. Develop a simple investment process.
  3. Design your reading priorities accordingly.
  4. Build relationships with people you trust.
  5. Be accountable for mistakes.

Just do your best, and let the chips fall where they may.

Your Feedback

Thanks for sticking around to the end. A complete version of the entire How to Read Financial News Redux series, with endnotes, is forthcoming. But until then, I welcome your comments.

  • What did I get right?
  • What did I miss?

If you liked this post, don’t forget to subscribe to the Enterprising Investor.

All posts are the opinion of the author. As such, they should not be construed as investment advice, nor do the opinions expressed necessarily reflect the views of CFA Institute or the author’s employer.

Image credit: ©Getty Images/Morsa Images

Continuing Education for CFA Institute Members

Select articles are eligible for continuing education (CE) credit. Record credits easily using the CFA Institute Members App, available on iOS and Android.

About the Author(s)
Robert J. Martorana, CFA

The late Robert J. Martorana, CFA, worked on the buy-side as a stock analyst, portfolio manager, research director, financial advisor, and editor of a hedge-fund website. In 2009, he founded Right Blend Investing, a fee-based RIA that managed individual portfolios and did consulting for the asset management industry. RBI has one unique claim to fame in that it supported an orphanage in Andhra Pradesh, India. Martorana published over 1,000 pages of contract research and co-authored Alts Democratized by Wiley Finance.

7 thoughts on “How to Read Financial News Redux: Prevent Memory Contamination”

  1. Nicholas F. Morris says:

    Robert, I believe in crystal balls and the ability to predict the past. However here you’ve brought the fragmented elements of behavioral finance together in a comprehensive fashion that is easily digested. Perception and relativity impact our decision making, but you included so much more the new investor needs to know before market exuberance overcomes sustainable resources. Hear no evil, see no evil, just follow the numbers.

    I look forward to the release of “How to Read Financial News Redux” series. Thank you!

    Best Regards,
    Nick M.

    1. Rob Martorana says:


      Thanks for your feedback.

      It seems that much of the academic effort in behavioral finance focuses on documenting biases instead of overcoming biases. After 34 years in finance I know that confirmation bias is pervasive, and I know it affects me personally. So my goal is to develop a reading process that minimizes its effects.


  2. Rob Martorana says:

    Clarification About the Word “Consensus”

    In this article I refer to consensus in two ways: consensus estimates and consensus opinion. A reader pointed out that this was confusing, and I realized that I was using two definitions for the same word:

    1) A consensus can mean a generally accepted opinion or belief. In the context of news, consensus means the accepted narrative or conventional thinking.
    2) A consensus also means an “average projected value,” such as a consensus forecast or consensus earnings estimates

    In this article I discussed the consensus forecast for 2019 corporate earnings. I refer to consensus estimates collected by FactSet, a dataset that averages brokerage estimates of corporate earnings.

    Generally speaking, consensus estimates for corporate earnings are considered reliable: It’s hard to come up with a more robust prediction than the wisdom of the crowd. In fact, some people believe that it’s not worth the effort to make independent estimates of corporate earnings since you are unlikely to improve on the accuracy of published estimates.

    Do Consensus Estimates Qualify as “News”?
    I am not sure how to categorize consensus estimates. Are consensus estimates a “fact” or merely an “opinion”? Earnings estimates are certainly a form of information, and there is strong evidence that earnings estimates affect stock prices. But I am not sure how to categorize this information. To the best of my knowledge, estimates are an ambiguous category of information, like opinion polls.

    My thanks to an astute reader for pointing this out.

  3. Sunil Kurmi says:

    Hello Rob Martorana,

    I really like your article the way you have describe the entire thing in such a descriptive way. I really like it.

    Hope you believe some more.

    Thank you

  4. Kirk Cornwell says:

    Part of this is not being as intimidated by job titles as the media sometimes seems to be. For example, a “Federal Reserve Governor” actually might not be the best person to listen to regarding the wisdom of a rate cut. Stock analysts usually have a buy-side bias. Interviewers can be caught up in the enthusiasm of CEOs. (“If it’s in the [Wall Street] Journal, it’s in the market” comes to mind.)

  5. Mike Czekaj says:

    Thank you for making this simplified, yet informative. I recently saw a post somewhere that said the following: trading costs to zero, data is cheap and available to everyone, data increases confidence, but doesn’t improve forecasts….the opportunity to be a terrible investor has never been better. Crude? Yes, but also some truth. I think you offer some great tips to get around this! Thanks again!

  6. Rob Martorana says:

    Thank you!

    Regarding the titles of a person, and their reliability as a source of information. Let me address this in two ways:

    1) Yes, you’re right, a person in authority is not always correct. This is why an “appeal to authority” does not have the final say in a logical argument.
    2) On the other hand, a person’s title IS relevant if they are commenting about something within their circle of competence. Warren Buffett can speak with authority about value investing, for example. I’m not saying he is perfect, but I give his opinion on this topic greater weight than most other opinions.

    What about when a person’s title or experience does NOT qualify them to speak with authority? This came up in a prior article when a reader asked about Vanessa Otero. The person said that she did not have experience in media or journalism.

    Why did I cite Otero’s work? I valued her opinion because I could objectively see her process and methodology, and her conclusions seem quite logical. So sometimes, if a news source is diligent and well documented, we do not need to rely on their title or experience. Here is a link to the question and my response:


    Good point.

    Trading costs are near zero and data is expanding towards infinity. (Note: Transaction costs may be zero, but there is still the bid/ask spread and market impact.)

    But you point out the more relevant problem: Our forecasts have not improved. This is especially true in capital markets, where greater data has not led to better forecasts or superior alpha. That’s because transparency and efficiency have risen, and market prices rapidly reflect news. (I believe in semi-strong market efficiency as described by Eugene Fama. Most of the time, anyway…)

    Even worse, our CONFIDENCE rises without bound! We have all of this data so we must be smart! Yup, that describes all of us, I’m afraid.

    Our pride can’t be eliminated, but it CAN be mitigated.

    Thank you!

Leave a Reply

Your email address will not be published. Required fields are marked *

By continuing to use the site, you agree to the use of cookies. more information

The cookie settings on this website are set to "allow cookies" to give you the best browsing experience possible. If you continue to use this website without changing your cookie settings or you click "Accept" below then you are consenting to this.