Active reading can even overcome sleepiness and result in comprehension

I know better than to put off reading my 1000 words a day from the geological literature until late I the day, yet I chose to do so anyway. Today the impulse to do so was triggered by Spring. There were very few people in our geology department today, as most people chose to leave for their Easter long weekend as early as they could. Some of the general staff were here, but were only scheduled to be here till about 2:00 pm. Around the time that they were leaving I noticed that it was a beautiful day, nice and sunny, yet not hot at all, and I felt restless. Therefore I decided to head out on an adventure, and went into the city center to admire the street artists and musicians who perform for the crowds there. I also stopped in the American Book store and picked up a new book in a favourite series.

After these adventures I managed to limit myself to only an hour of reading in the new book before crossing the street to my office again around quarter to nine, intending to finish the work I’d interrupted hours before. Instead I procrastinated by catching up on reading blogs/e-mail/etc, updated my financial records to show today’s spending, and chatted with a friend on line. As a result I didn’t start reading my 1000 till 11:30 pm, and it was difficult to find the discipline I needed to actually complete the process.

However, hard as it was to make myself do, I am now a total convert to this “active reading” thing. The need to type the below notes meant that I had to actually pay attention to what I was reading, and not just sleepily skim over the words catching one in ten. I am positive that focusing on each paragraph one at a time and typing the notes as I go is the only thing that made it possible to understand what I was reading tonight. Perhaps I can understand and retain what I read the “normal” way when I’m high in energy, but this late in the day “active” reading is required. So, without further adieu, I give you the second-to-last installment of my paragraph-by-paragraph summary of the paper I’ve been reading (see a few days back for the citation and the link to the article itself).

The next section of the paper is on the 40Ar/39Ar results.

Paragraph 63 introduces the two pegmatites from which muscovite was obtained for the dating, describes the location of the pegmatites, gives the age results (~47.1 Ma for one and ~46.9 for the other). They state that the samples were obtained from above and below the contact, and are within error of one another. They state that this means that these numbers represent the time of cooling, rather than crystallization. This ends the section on the dating, and also concludes the section on their results. The next section is the Discussion.

Paragraph 64 points out that previous studies described a fault in this area and described the rocks on either side as high-grade on the west and low-grade the east, but this study indicates that what had been called “low grade” has actually undergone sufficient burial to achieve >650 C at 8-10 kbars, or largely the same as the “high grade” Gneiss. They go on to mention that the microstructures near the contact between the two units reveal progressive deformation & a high-T constrictional shear zone between the units, & subsequent overprinting by low-T deformation. They say that therefore this area is complex & important in context with the entire Cascades region, and for strike-slip dynamics in orogeny.

This concludes the lead-in portion of the discussion; the next section is on the Previous interpretations of Ruby Mt-Elijah Ridge tectonic history.

Paragraph 65 names three different studies which have addressed the tectonic history of this area. The first called this area a suture of Insular and Intermontane belts since it puts ocean rocks structurally above continental arc rocks. The second calls it a tectonized intrusive contact without a through-going fault, saying tilting is enough to expose the largely intact Mid-Cretaceous section, and add in one more fault to account for the change in pressure across the area. The third described this area as a major tectonic boundary, since there are mylonites present in addition to the difference in metamorphic pressures.

Paragraph 66 gives one other possible interpretation (calling this area the uppermost part of the flowing orogenic crust with the Napeequea unit acting as a rigid lid of a layered crust. They describe this interpretation as analogous to migmatite-upper crust relationships observed in the hinterland of the orogen, and cite a source (but from the sentence itself it isn’t clear if the source proposes this interpretation or simply describes the relationships in the analogy). They then state that all four models can be considered in light of their new data. This concludes the section on previous interpretations. The next section is the Re-evaluation of Ruby Mt-Elijah Ridge tectonometamorphic history.

Paragraph 67 describes the units in this area (orthogneiss dominating, Napeequea over it on Ruby Mt. and part (west end) of Elihah ridge, and Methow rocks above it on the east part), the dip (35-45 to the east) of the planar fabrics of bother Napeequea rocks and orthogneiss on Elijah Ridge while the far east side of the ridge is folded (= different deformation style across the ridge).

Paragraph 68 specifics that the low-T fabric overprinting migmatite seen in the microstructural analysis occurs only in the structurally deepest part of the exposed orthogneiss at the base of Ruby Mt. the only other units to show such overprint is the tonalitic amphibolite at the Skagit Gneiss-Napeequea contact .

Paragraph 69 contains too much information for one paragraph. It states that the strain gradient near the summit of Ruby Mt. displays more constrictional fabrics near the contact between the Skagit orthogneiss and Napeequea units (both of which are L-tectonites). It mentions the pegmatite which intrudes both that was deformed and cooled at ~47 Ma. It lists the clues which lead them to conclude that the shear zone is a medium- to high-T feature. It acknowledges some lower T-overprinting in a few of the units. It suggests rheologic contrast as the cause of the shear zone, but doesn’t decide between shear zone formation during or after emplacement of the orthogneiss. It suggests that the shear zone is an old contact which preserves the early high-T deformation of both units, further suggesting that the deformation either developed along or transposed the intrusive contact.

Paragraph 70 mentions the contacts between these units in other areas, the fact that the quartzite and schist of the Napeequa unit is generally more deformed than is the orthogneiss, suggesting that the composition of each is likely responsible for that difference. It then states that the contact exposed on Ruby Mt. is the one showing the most intense high strain fabrics.

This is not a section break. However, there are no more section breaks in this paper, save for the one which precedes the final paragraph, and I’m at 1000 words read today, and 1000 more left before the end of the paper. Therefore I’m going to stop here and pick it up with the next paragraph tomorrow.

getting faster

I just read and took notes on the next section of the paper I’ve been “actively reading” for my 1000 words a day from the geologic literature, and am pleased to note that while I read well over 1000 (bringing yesterday’s and today’s combined total over 2000 words), it took just under 35 minutes to do so. I don’t know if this is because I’m becoming more comfortable with the process of active reading and typing up notes on each paragraph I read as I read them, or if these notes are less extensive than those for the earlier section, but I’m hopping it is just a matter of practice making me more efficient. I am willing to make 30 minutes a day, every day for this goal, but the over an hour I spent the first time I tried this technique might be asking a bit much.

I don’t know if anyone is actually reading the notes I’ve been posting, but I shall continue to post them till I reach the end of this article. If it should happen that someone tells me that the notes are interesting and/or helpful, I could easily be talked into continuing to post them for future articles I read, too. However, I suspect that for most people who are looking for interesting blogs to read one article summarized one paragraph at a time will be plenty. Feel free to let me know if I’m mistaken on that point.

Today’s notes are from the section on microstructural analysis:
Paragraph 53 introduces the locations from which the samples were collected for microstructural analysis and gives the goals for this portion of the study (understand mechanisms and relative timing of juxtaposition of the units)
Paragraph 54 mentions which samples were chosen for electron back scatter diffraction, gives details of what was found and compares and contrasts the results from the leucosome and mesosome layers of the sample. They go on to offer a tentative (since there weren’t enough samples analyzed to be confidant) interpretation (possible late shear zone affected the structurally deepest exposed rocks)
Paragraph 55 names a sample chosen for EBSD, reiterates that it is from the Skagit Gneiss), gives published U/Pb zircon age for this sample (which is younger than for other Skagit Gneiss samples), and describes the areas in this sample for which they made EBSD maps.
Paragraph 56 lists the results of the EBSD analysis for this sample (including strong crystallographic preferred orientation, prism slip for quartz (both regions of sample), sub grain rotation and recrystallisation (in shear zone), different grain sizes in different regions.
Paragraph 57 describes an effect (some qtz not deformed adjacent to larg plg grain) which only shows up in EBSD map, not via traditional optical methods. It also points out that such variety in qtz textures could indicate qtz deformed at all T and may have preserved evidence for more than one condition.
Paragraph 58 gives EBSD results for a new sample, from a new location.
Paragraph 59 mentions that the above sample has overprinting which doesn’t show up in higher parts of Ruby Mt. (instead the higher parts preserve high-T fabric with not much qtz recrystalization).
Paragraph 60 discusses textures in the structurally highest regions of Skagit Gneiss, and interpret them to indicate that the shear zone is a high-T feature.
Paragraph 61 more EBSD results this time for another couple of samples, comparing and contrasting them with each other, and with the other samples in the area. Interprets the results to indicate that the quartzite may have taken up much of the low-T deformation in this area
Paragraph 62 gives the EBSD results for samples from the RLFZ because they feel that understanding deformation in the fault zone will help interpretation of the role of the fault in burial and exhumation. They point out the similarities between these samples and the above.
Here ends the microstructural analysis section. I found it slightly hard to follow in terms of understanding which samples/areas were being discussed when, but that was because I hadn’t made notes about the sample names/locations in my pre-reading familiarization session. Nonetheless, I think that sub headings might have helped.
The next section will cover the 40Ar/39Ar results.

Quality is better than quantity

The section of the paper I read today is just under 1000 words, but the next section of the paper is a long one—if I read on it will be over 2000 words of reading for the day. While that is not necessarily a bad thing in and of itself, I do have other things I need to do with my day, so I will choose to cut today’s total # of words a bit short, knowing that today and tomorrow will combine to meet the goal of reading “1000 words a day”. It is the sprit of the law which is more important than the letter in this case, and I feel that the “active reading” I’ve been doing, summarizing the contents of each paragraph as I read it, and before going on to the next paragraph is more than good enough in terms of quality of reading comprehension and retention to make it reasonable to break at a section break, rather than at an exact word count.
So, without further adieu, here follows my summary of the paper I’m reading. See the past few days for the earlier sections, and stay tuned tomorrow for the next section.
The next section covers Thermometry and Barometry. They present thermometry first.
Paragraph 44 lists the three techniques they used to calculate metamorphic temperatures. They specify that unzoned garnet and matrix biotite compositions were used for one set of calculations, that garnet core + green matrix hornblende and garnet rim + blue-green hornblende was used for the other, and they don’t specify here what minerals were used for the Thermocalc compositions.
Paragraph 45 reports temperatures of 700-800◦ C for the grt-bt schists of Ruby Mt. and 675-730◦ C for the grt-hbl in the amphibolites. They say that the amphibolites above and below the contact between the Skagit Gneiss and the Naeequea give results within error of one another.
Paragraph 46 reports for Elijah ridge 650-700◦ C for both ky-st & and-crd schist and 570-670◦ C for the amphibolite. They specify that Elijah Ridge gives lower temps than Ruby Mt., but point out that errors are ~50◦ C so it is hard to say if difference is real. They also point out that the andalusite is a late phase, so the T pre-dates that mineral’s growth.
That concludes the thermometry section, the next section is barometry
Paragraph 47 lists the three barometers used, and specifies that garnet core and matrix plag gave max P. They also state that they usually assumed that kyanite is the stable Al2SiO5 polymorph, and named the one exception to that.
Paragraph 48 states that most samples from this study yield pressure estimates of 8-10 kbar, which agrees with previous studies in the area. They state that while Elijah Ridge gave slightly lower results, they are still within error of the others.
Paragraph 49 addresses garnet zoning, pointing out that only along the eastern margin near the RLFZ can one find garnet that has discontinuous zoning in the grossular component (see pretty maps of fig 8)—they state that this pattern could relate to GASP reaction of coexisting grt and plg, in which case the increase in Ca towards garnet rims equals an increase in pressure as they grew. They go on to specify that structurally deeper samples have homogeneous garnet or only a thin retrograde rim zoning. They further state that the above mentioned zoned(Ca) garnet eastern margin samples saw lower T than the ones with homogenous garnets, and say that the zoned ones might be a better record of the prograde history and they also seem to have a more complex later history given magmatism and faulting associated with RLFZ. (This paragraph was the hardest to summarize yet, being packed full of lots of information and not flowing from one thought to the next as well as it might.)
This ends the section on Barometry. The combined section on Thermometry and Barometry includes a figure showing all of the above P/T results—the individual error squares overlap from one to the next, giving a range that is visually continuous, despite the fact that the lowest and highest individual samples don’t overlap. The next section is on P-T paths.
Paragraph 50 mentions three lines of evidence for different stages of the PT path: 1) evidence for pre-andalusite conditions for the andalusite-cordierite schist with 2) garnet rims showing increase in P for those rocks. 3) the fact that kyanite used to be present in all samples, so max P had to be in kyanite zone. They go on to state that the max P/T reported in last section was followed by near isothermal decompression to less than 5 kbars while T was still high.
Paragraph 51 lists textures which point to metamorphic reactions continuing during decompression. Two of these textures are observed in this study, and they mention that their observations are in agreement with a previous study, and list other lines of evidence from that study.
Paragraph 52 states that the P-T results from the eastern side of Elijah ridge are for conditions which preceded a low-P metamorphic overprint. It also equates the P with depths of ~25-33 km, which is equivalent to results from farther to the west in Skagit Gneiss.
This ends the P-T paths section. Their proposed path focuses on the max P/T conditions followed by decompression while T was still high. The next section is on microstructural analysis.

Yes, I think the extra time is worth it

I find that I’m liking this “active reading” stuff. It used to be that on days like yesterday, when I didn’t get around to reading my 1000 words of geologic literature till very late in the day I’d “read” it, as in my eyes would look at each word on the page, but it was more along the lines of “skimming” it, and it is somewhat doubtful how much, if any, was actually retained. Yesterday, however, I actually got meaning out of each and every paragraph I read. I know this, because it isn’t possible for me to type a sentence or three about what it says if I don’t understand it first.
Normally I’m a two-monitor kind of person. If I need to read something and type notes about it, I keep the article on one screen, and the word-processor on the other. Yesterday, on the other hand, that wasn’t possible. I’d spent the day using the electron microprobe and when it was time for the ‘probe’ operator to head home for the day I was tired, so I simply took my computer home with me, rather than bringing it back up the stairs to my office. As a result when I realized hours later that I still needed to do my 1000 for the day (I’m at 82 days in a row now this time!) I was faced with the choice of dealing without a second monitor or carrying the computer back to my office.
The solution I hit upon worked very well—I’d copy-paste a single paragraph into the document in which I was working, read it, type up my summary, then move that paragraph into an otherwise empty document. At each minor section break I’d do a word-count of the document which contained only the paragraphs I’d already read, to see if I’d done 1000 yet. As it turns out, I hit 1053 words at one of the minor section breaks, to the notes below don’t end in quite as logical of a place as the previous batch, but I was tired (Europe only just did the change to daylight savings this weekend), so called it good. It is interesting to note that my summary, including the headings “Paragraph 1, 2, etc.” took 581 words to summarize the 1053 words of the original.
When first I started reading 1000 words a day I thought of it as a 20-minute time commitment. Reading this 1053 words and typing up a summary took 54 minutes. Therefore it takes nearly three times as many minutes, but I think that the gain is more than four-fold the amount of information retained.
Here follows the summary in progress for the article Metamorphism and deformation at different structural levels in a strike-slip fault zone, Ross Lake fault, North Cascades, USA by Gordon et al 2010:
Paragraph 18 informs us that it can be difficult to find appropriate mineral assemblages for P-T estimates in the Skagit Gneiss. They mention a previous study which used 4 metapelites, and tell us that this study found 4 useful metapelites and three garnet amphibolites between the Skagit Gneiss and the Napeequa unit.
The next section focuses on the garnet amphibolites
Paragraph 19 specifies the locations for the three garnet amphibolites that yielded PT estimates and gives the units thereof.
Paragraph 20 mentions which two amphibolites (from Ruby Mt.) are migmatic adn which one isn’t (from Elijah Ridge) and gives the mineral assemblages and fabric.
Paragraph 21 describes the garnet in one of the amphibolites, giving size, general composition, mentions that it isn’t zoned (and states that this is normal for the unit), list of inclusions, and the fact that some of them display post-kinematic coronas.
Paragraph 22 describes the garnet from one of the two other samples in the other unit, mentioning their size (smaller than last paragraph), and the fact that minor growth zoning is present, and points out that the final sample is very similar to this one.
Paragraph 23 describes the hornblende of all three samples, two of which have zoning (just like the garnet from the same unit), but zoning is rare in the other sample (just like the garnet from that sample)
Paragraph 24 mentions the zoning of the plagioclase in all three samples, states that for the Ruby amphibolites the reverse (An increase to rim) zoning is more common & more variable than the normal (An decrease to rim) zoning that is present.
Paragraph 25 describes the plagioclase zoning for the Elijah ridge amphibolite, which also has more reverse than normally zoned examples. However, in this case the normal zoning is more variable than is the reverse.
Paragraph 26 mentions which sample contains clinopyroxene, and the fact that it isn’t zoned and is both in the matrix and as inclusions within the garnet.
Here ends the section on the minerals present in the amphibolites. The next section looks at the metapelites.
Paragraph 27 lists the locations (2 from Ruby Mt, and 2 from Elijah Ridge) and general assemblages of the four metapelites used in this study, and specifies which ones are structurally higher than their neighbors, and which pair is structurally higher than the other.
The next section focuses on the Ruby Mt. Metapelites.
Paragraph 28 lists the major (grt-bt-sil-ky-crd-qtz-pl) and accessory (il-zr-apt-mnz) minerals present in these rocks and point out that the sillimanite one has pseudomorphic textures hinting at former kyanite while the kyanite bearing one has some sillimanite. It also specifies the habits of the cordierite
Paragraph 29 describes the fabrics present, and where possible mentions what that says about the T at which each deformation happened.
Paragraph 30 describes the habits of garnet in one of the samples (including what is included therein) and cordierite and points out evidence for a retrograde reaction of grt-crd.
Paragraph 31 continues with the same sample as last paragraph,stating that plagioclase isn’t generally zoned, the biotite is homogeneous, and describes alignment of sillimanite with the foliation. It also reminds us that kyanite used to be stable in this sample.
Paragraph 32 moves on to the other sample, lists minerals present, compares garnet, plag, and biotite with last sample, describes the kyanite and sillimanite present.
There ends the description of the minerals in the metapelites from Ruby Mountain. The next section will describe those from Elijah Ridge.

my first attempt at actually doing the “active reading” technique—part one: look over the paper without reading the details

Since I am an addicted reader of novels, the suggestions for active reading that I summarized in my last post sounded tedious and cumbersome to me when our teacher suggested them. Therefore I decided to work through one paper using those techniques here on this blog, because if I do it publicly, I will feel obligated to continue the process all the way through to the end, rather than giving up and returning to lazy reading techniques. If, at the end of this process it feels like the extra effort has been worth it, I shall let you know.
Today’s paper is Gordon et al. (2010)*. This paper was chosen for today’s reading because my PhD advisor suggested it and another article from the same issue of the journal as having structures similar to what I should be aiming at when I write the paper summarizing what I did for my PhD research. Therefore the questions I will keep in mind when reading this paper are:

1) What structure did they choose, and how effective do I feel it is in presenting their work? 2) How many different techniques did they use for this project, and how did they organize their presentation of these various techniques? 3) How effective is that organizational method? 4) Would another order be more appropriate? 5) Do they indicate how much more work they undertook in addition to the specific results they share? 6) Is there any indication as to how they selected these specific results? 7) How do their results and the presentation thereof lead to the conclusions they draw? 8) Do their results and the subsequent discussion convince me that their conclusions are appropriate? 9) Why or why not?

Step one in the active reading is simply to look over the entire thing. This is a 20 page document, the first 17 of which contain text, tables, and figures. The headings used in this article reduce to the following outline:

INTRODUCTION
GEOLOGICAL OVERVIEW OF THE SKAGIT GNEISS AND ROSS LAKE FAULT
. Eastern margin of the Skagit Gniess
. Ross Lake Fault Zone
. Ruby Mountain and Elijah Ridge
PETROGRAPHY AND MINERAL CHEMISTRY
. Garnet amphibolite
. Metapelitic rocks
. . Ruby Mountain Metapelites
. . Elijah Ridge Metapelites
THERMOMETRY AND BAROMETRY
. Thermometry
. Barometry
P-T PATHS
MICROSTRUCTURAL ANALYSIS
40Ar/39Ar RESULTS
DISCUSSION
. Previous interpretations of Ruby Mt-Elijah Ridge tectonic history
. Re-evaluation of Ruby Mt Elijah Ridge tectonometamorphic history

The article contains 13 figures.
Figure 1 contains both a simplified geologic map of the region of study area and an annotated Google Earth image (the latter is in colour in the pdf version). The notes include details of sample locations, locations of geological units, and pressure temperature estimates.
Figure 2 is a closer-scale geologic map of just the fault zone region
Figure 3 shows a series of cross-sections across the region.
Figure 4 contains colour photomicrograph of thin sections of two samples
Figure 5 contains colour photomicrograph of thin sections of three samples
Figure 6 contains colour photomicrograph of thin sections of three samples
Figure 7 contains both a colour photomicrograph of one thin section and major element x-ray maps for the garnet in that thin section for Mn, Fe, and Mg (this garnet is obviously much richer in Fe than Mg or Mn)
Figure 8 contains both a colour photomicrograph of another thin section and major element x-ray maps for the garnet in that thin section for Ca, Mn and Fe, (this garnet has a core which is Ca-poor and a rim (nearly as thick as the core) which is Ca-rich—the core is richer in Mn than the rim. Both the core and rim contain a similar amount of Fe overall, but there is an Fe-poor region at the core-rim boundary, which could be related to the greater quantity of inclusions in that region (my observation, not what the caption said))
Figure 9 is a P-T diagram showing the estimates obtained for various samples plotting in a clump just above the ky-sil boundary in the range 600-800 C and 7-11 kbars. It also includes arrows to show the near-isothermal decompression they infer for these samples based upon mineral assemblages and textures.
Figure 10 contains two (colour) photomicrographs of the orthogneiss, showing two different grain sizes.
Figure 11 contains field photos illustrating intense constrictional fabrics to two different rock types from the region.
Figure 12 contains Muscovite 40Ar/39Ar age spectra for two different samples.
Figure 13 contains a map-view cartoon sketch of the transpressional step-over and duplex structures of the region.

Of the questions I asked myself at the beginning, I feel that just looking at the headings and figures permits me to answer the following:

1) They present an introduction and regional geology, followed by six different sections presenting each of their different types of results, followed by a discussion section where they first list previous interpretations of the area and then share how they feel their new data modifies those interpretations (Isn’t it funny how I can state that with confidence, even though I have yet to actually read the paper and so have no idea what their interpretation is, nor what the older interpretations might have been?)
2) The six techniques they used are geological mapping, petrography/mineral chemistry analysis, thermobarometric calculations, P-T paths, microstructural analysis, and geochronology, listed in that order. The order may well have been chosen because one must first do the field work and obtain the samples before anything else happens. The microstructural section could just as easily have come before the petrography/mineral chemistry section, but that work is essential for the geothermobarometric calculations, which, in turn, is essential to determine a P-T path. The age dating could have been presented at any point after the field work, unless 40Ar/39Ar technique requires information obtained in the petrography/mineral chemistry section.
3), 4) How effective is that organizational method? The organizational method looks logical thus far, I’d have to actually read the paper to see if it is truly effective and if another would have been more appropriate.

The remaining questions can’t be answered till I read the text itself. This post is now quite long enough, so I will take a break from “reading” this paper and share with you my progress to this point. Stay tuned for my paragraph by paragraph summaries of this paper, once I’ve done them.


*Gordon, S.M., Whitney, D.L., Miller, R.B., McLean, N., and Seaton, N.C.A., 2010, Metamorphism and deformation at different structural levels in a strike-slip fault zone, Ross Lake fault, North Cascades, USA: Journal of Metamorphic Geology, 28, 117-136.

Active reading techniques to improve understanding & retention

I spent this week taking a short course on Successful Scientific Writing: from Proposal to Publication taught by Dr. Sarah Shephard at the Center for Teaching and Learning at ETH Zürich. The course provided such a wealth of useful information I’ll try to do several posts about the things I’ve learned. Today’s post will be all from memory, as I left the notebook full of handouts at home when I came into the office this morning.

Today’s topic is reading. Reading is something at which I’ve always excelled—I can’t remember the time before I was able to read, and I’ve always been one to “fall into” books, getting totally immersed in the story and not noticing time elapsing in the real world. Unfortunately, reading articles in scientific journals does not have that same effect. Narrative stories are designed to flow smoothly and to engage the audience, but most science writing is designed to communicate results of a project and discuss the implications thereof. We read journal articles with the goal of learning, not entertainment (though, in some fortunate cases both are possible), and she recommends a very different approach to reading science literature than is taken when reading a novel.

She suggests first reading the literature when you are in the “deciding upon a project to do” stage—find out what is the state of the art on the topic(s) of interest, not only what is currently known, but how those facts have been interpreted thus far, and if there is more than one “school of thought” on the topic. Which author(s) do you agree or disagree with, and why? Keep these questions in mind when doing the initial reading of the literature.

When you sit down to read a new article, first glance over the entire article—make a note of the structure of the article—how much of it is devoted to the introduction, the methods used, the results, and the discussion/conclusions? What headings did they use for the various sections? What figures and tables were used to illustrate their points or provide supplementary data? Once you are comfortable with what information is going to be presented think of specific questions you expect to be answered in the article, and make a note of them. Now you are ready to begin the reading process.

The reading itself should be punctuated by writing on the part of the reader. Read one paragraph, and then jot down notes to yourself about what it said. Paraphrase their point in your own words—this is an important component of learning the subject matter—by taking an active role in the process and writing it down, you will better remember what it said. After you have paraphrased the paragraph ask yourself the following questions: Did it provide the answers to any of the questions on your list? Did it make you think of other questions you wish to find answers to within the course of the article? After you have completed all of these tasks read the next paragraph and repeat the process of writing a paraphrased summary and determining if it has answered any of your questions or posed new questions. Once you’ve done this process for a number of paragraphs (five is a good number, but there may be reasons within the structure of the article you are reading to include more or fewer) go back and look over all five paragraphs and your notes thereon and determine what combined information they relate—write down your own summary of the whole section. Ask yourself if this section answers any of the questions you have. Do you agree or disagree with what has been said thus far and why? Does this section of the paper agree or disagree with other papers you have read? Repeat this process through the entire paper, and at the end condense it all into a brief summary of what you feel the key points are relevant to the topic you are currently studying.

She cautions that this process takes longer than simply reading each word on the page in order one time through. However, she also insists that it is worth the extra effort on your part because these techniques greatly enhance one’s ability to understand and remember what one has just read. She also says that not only should one read everything one can find that is relevant to the topic at the start of the project before doing the new research, one should also go back and review all of those papers (and any new ones published subsequently) after obtaining one’s results and before actually writing the paper to publish the results. That second reading should go much faster, as you will have your old notes on the paper to look at, in addition to the paper itself. However, having done the research you may find that you now have different questions or feel that different aspects of each paper is now more important than the parts you had initially recorded.

Because this reading method is more intensive than simply “falling into a good book” she recommends that it be done in small segments—read for 10 minutes at a time, then take a short break. Someone who is experienced in this technique might be able to read (and do the writing/thinking about what has been read) for 30 minutes at a time before needing a break. However, remember to make the break short and return to the process!